Effect of auxiliary cathode on plasma nitriding behavior of Cr12MoV alloy steel

Samples with dimensions of 20 mm × 10 mm × 2 mm were cut off from a Cr12MoV alloy steel plate; their chemical composition is shown in table 1. All samples were polished using 1000-grit SiC paper and then ultrasonically cleaned in acetone for nitriding.

The plasma nitriding was carried out in a DLZ-10/10/20 semi-industrial pulsed-DC plasma nitriding apparatus, shown schematically in figure 1(a). The apparatus consisted of a 1 kHz square wave pulsed-DC power supply and a heating system to increase the temperature up to 1000 °C by radiation. The sample was attached to a carbon steel cathode disc by a Cr12MoV alloy steel holder with a thickness of 10 mm for nitriding. At the same time, a Cr12MoV alloy steel plate with dimensions of 40 mm × 40 mm × 5 mm was placed opposite to the sample as an auxiliary cathode to form an HC. After the chamber was evacuated to a pressure lower than 3 × 10⁻¹ Pa and subsequently filled with argon, the sample was heated to the temperature over 300 °C at a pressure of 200 Pa and cleaned by argon sputtering for 15 min at a pressure of 60 Pa using a cathode voltage of −900 V with duty cycle of 10%. Then, the atmosphere was switched to ammonia gas. The samples were nitrided for 4 h at a pressure of 100 Pa, voltage of −650 V, duty cycle of 50%, and temperature of 550 °C. A WDL-31 infrared thermometer was used to measure the temperature of the sample.

Zoom In Zoom Out Reset image size

When the auxiliary cathode is positioned at an appropriate distance (d) from the sample, HCD occurs between them as shown clearly in figure 1(b). This promotes the nitridation of the sample in the case when the above nitriding parameters were not varied. To investigate the effect of the auxiliary cathode on the plasma nitriding, the Cr12MoV samples were nitrided at distances d in the range of 6–24 mm. For comparison, the sample was also nitrided without the assistance of the auxiliary cathode.

The morphologies and microstructures of the alloy steel samples before and after nitriding were investigated using optical microscopy (OM), x-ray diffraction (XRD) and grazing incidence x-ray diffraction (GIXRD), field-emission scanning electron microscope (FE-SEM)/x-ray energy-dispersive spectroscopy (EDS), electron probe micro-analyzer (EPMA), and transmission electron microscopy (TEM). The samples for OM analysis were etched in 4% nitric acid alcohol solution. Microhardness tests were conducted using a HV-1000 instrument with a load of 50 g and dwell time of 10 s. At least five readings were taken to obtain each average hardness value.

The XRD pattern shown in figure 2(a) indicates that the Cr12MoV alloy steel consists of α-Fe matrix and Cr₇C₃-type carbides. These carbides, which appear as bright areas under OM, nearly form a carbide network, as shown in figure 2(b). They exhibit a grayer contrast in figure 2(c), a SEM image in backscattered electron (BSE) mode. It shows that the sizes of the carbides are in the range of one micron to dozens of microns. The EDS analysis indicates that the carbides contain major Fe and Cr and minor V.

Zoom In Zoom Out Reset image size

Figure 3 shows the GIXRD results of the Cr12MoV alloy steel nitrided for 4 h without and with the auxiliary cathode at a distance d = 8 mm from the sample. The top surface of the alloy steel nitrided without the auxiliary cathode is composed of major γ'-Fe₄N, CrN, and minor α-Fe(N) phases. When using the auxiliary cathode during the nitriding process, major ε-Fe_2–3N and minor γ'-Fe₄N phases appear, which indicate that the auxiliary cathode assistance promotes the nitridation of the alloy steel.

Zoom In Zoom Out Reset image size

After 4 h of nitridation without auxiliary cathode, the Cr12MoV steel forms a nitrided surface layer with a thickness of ∼40 μm. The etched nitrided layer exhibits darker contrast under OM as shown in figure 4(a). Figure 4(b) shows a BSE-SEM image of the nitrided layer in a higher magnification. The appearance of large particles in deeper areas is the same in grayscale as that of the original carbides (figure 2(c)), while those in a shallow depth developed a deeper gray contrast due to nitridation. This indicates that the original large carbide particles were fully nitrided in the shallow area but not in the deeper. Figure 4(c) shows a magnified view of the framed area in figure 4(b), indicating that the periphery of large carbide particles, which, apparently were not been nitrided, is converted into nitrides, resulting in the deeper gray contrast as indicated by the arrows. However, the fine carbide precipitates of around and below 1 μm in the ∼40-μm-thick nitrided layer, were converted to deeper gray particles, suggesting that they were nitrided.

Zoom In Zoom Out Reset image size

Figure 5(a) shows an OM view of the cross-sectional microstructures of the etched Cr12MoV alloy steel after nitridation for 4 h with the application of the auxiliary cathode (d = 8 mm). The auxiliary cathode assistance results not only an increase in the thickness of the nitrided layer from ∼40 μm to ∼80 μm, but it also changes its microstructure, which lies in two aspects. First, a top compound layer with a thickness of 10.8 μm occurs, which can be identified as mainly ε-Fe_2–3N by GIXRD characterization. Second, the deeper area exhibits certain new 'veins' with a color contrast similar to that of the surface nitrided layer. These features of the compound layer and vein-like phases can be seen more clearly in figure 5(b), which shows the BSE-SEM image of the nitrided layer in higher magnification. Moreover, different from the nitridation without auxiliary cathode, apparently the original large carbide precipitates were nitrided, forming nitrides with deeper gray contrast.

Zoom In Zoom Out Reset image size

Figure 6 shows a comparison of the backscattered electron image (BEI) and corresponding x-ray elemental mappings of the Cr12MoV alloy steel nitrided without and with auxiliary cathode at a distance of d = 8 mm from the sample. A ∼40-μm-thick nitrided layer rich in N is visible on the alloy steel nitrided without auxiliary cathode (figure 6(a)). In addition, there is an area richer in N in the nitrided layer, as indicated by an arrow. This is undoubtedly due to the nitridation of the large Cr₇C₃-type carbide precipitates in shallow depth. No N signals were observed from the carbide precipitates below the nitrided layer, which implies that the precipitates in the deeper area were not nitrided. Under the assistance of the auxiliary cathode, the nitrided sample formed a much thicker (∼80 μm) nitrided layer, within which a thin top layer formed richer in N (figure 6(b)). The top layer nearly contains only N and Fe, in agreement with the GIXRD characterization indicating that major ε-Fe_2–3N formed there. Below the ε-Fe_2–3N layer, a number of veins rich in C appear. The veins result from the decomposition of large carbide precipitates by nitridation, which is detailed later. The comparison of results suggests that the nitridation behavior of the Cr12MoV became different under assistance of the auxiliary cathode.

Zoom In Zoom Out Reset image size

For the better understanding of the effect of the auxiliary cathode on nitridation, the microstructure of the nitrided layer below the top compound layer was further characterized by TEM. Figure 7 shows a TEM bright-field (BF) image of such region, in which the phase compositions on the specific locations of 1, 2, and 3 were identified by selected area electron diffraction patterns (SAEDP). The formation of Cr₂N (figure 7(b)) at location 1, suggests the presence of the original carbide precipitate. Cr₂N is formed due to incomplete nitridation of the carbide precipitates in larger size or in deeper area. This nitridation releases numerous C atoms, which diffuse preferentially along the grain boundaries where they react with Fe, precipitating veins of Fe₃C on location 2 (figure 7(c)). The major area around the Cr₂N particles and Fe₃C veins, such as in location 3, is α-Fe(N) phase (figure 7(d)). Its formation results from the penetration of a small amount of N into the ferrite phase of the alloy steel.

Zoom In Zoom Out Reset image size

The auxiliary cathode-assisted promotion of the plasma nitriding kinetics significantly affects the hardening of the Cr12MoV, as can be seen from the microhardness depth profiles shown in figure 8. For the alloy steel nitrided without the auxiliary cathode, the microhardness decreased slowly from 840 HV at the surface to 770 HV at the depth of 20 μm and then decreased rapidly to 287 HV at the depth of 60 μm, which is a value close to the hardness of the area without nitridation. In contrast, the alloy steel nitrided with the auxiliary cathode (at d = 8 mm) became harder. Its surface microhardness reached 1270 HV. The value slowly decreased to 923 HV at the depth of 40 μm and then quickly reached the hardness of the unnitrided area at the depth of 80 μm.

Zoom In Zoom Out Reset image size

The nitriding process of the Cr12MoV alloy steel without the auxiliary cathode can be described by a well-known model [1, 25]. The glow discharge decomposition of ammonia first occurs, generating N and H ions, which are accelerated by the electric field to bombard the alloy steel surface, sputtering Fe atoms. The Fe atoms are then combined with the active N atoms in the plasma, forming FeN species. A part of them are subsequently backscattered toward the sample and partly deposited on it, where they are decomposed to release N atoms. The inward penetration of the N atoms consequently forms the nitrided layer.

In contrast, the enhanced ammonia glow discharge process by the auxiliary cathode increases the concentration of the N active atoms, i.e., the nitrogen potential for nitriding. This results in the formation of a compound layer, which has not been observed in Cr12MoV alloy steels nitrided without auxiliary cathode. In that case, it appears mainly as γ'-Fe₄N, which is a nitride with lower molar concentration of nitrogen. At the same time, the high nitrogen potential results in the nitridation of large Cr₇C₃-type carbide precipitates in the alloy steel, releasing a number of C atoms. Then, they diffuse along the grain boundaries where they react with Fe to form Fe₃C 'lines'. The lines become waved vein-like, possibly because their formation is affected by the compressive strains generated during the nitridation.

The auxiliary cathode effect is closely related to the distance d between the auxiliary cathode and the Cr12MoV alloy steel. As can be seen in figure 5(a), an ∼80-μm-thick nitrided layer with a 10.8-μm-thick compound layer on its top was formed on the alloy steel at d = 8 mm. When d decreased from 8 to 6 mm, no significant nitrided layer could be observed by OM under the same magnification (figure 9(a)). A higher magnification under SEM in the BSE mode (the inset in figure 9(a)) reveals that only the carbide particles located near the sample surface were partially nitrided and consequently converted into a darker gray color contrast. When d increased from 8 mm, no significant variation could be seen in the thickness of the nitrided layer; however, the thickness of the compound layer decreased from 10.8 to 8.7 μm at d = 10 mm (figure 9(b)) and further to 4.9 μm at d = 24 mm (figure 9(c)). The dependence of the thickness of the compound layer on d is shown in figure 10. It is assumed that the ε-Fe_2–3N phase forms when nitrogen concentration exceeds the solid solution limit of the steel substrate during the nitriding process. Once the ε-Fe_2–3N phase develops a continuous layer, the nitridation process is controlled by the diffusion coefficient of nitrogen within the nitride layer. The diffusion coefficient of nitrogen depends highly on the nitrogen potential, i.e., the plasma density in the HC. As can be seen in figure 10, the highest plasma density was formed at d = 8 mm. As d increased from 8 to 24 mm, the plasma density slowly decreased; however, when d decreased from 8 to 6 mm, the plasma density sharply decreased. The dependence of the effect of the auxiliary cathode on d can be interpreted as follows.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Formation of the compound layer in the case when d varies from 8 to 24 mm indicates the generation of high-density plasma, which indicates that HCD occurred. An indirect evidence is the significantly higher glow brightness in the HC under these conditions than that in the case without the auxiliary cathode, even though the brightness decreases with the increase in d. During the HCD-assisted plasma nitriding process, the same negative potential was applied to the auxiliary cathode and the sample. Several electrons oscillate between repelling potential of the sheaths at opposite walls in the cathode (the auxiliary cathode and the steel). Compared with the conventional glow discharge between parallel plates, the number of ionization and excitation events of gas atoms due to electron collision is significantly increased in the HC (the so-called pendulum effect [26–29]). In particular, a large number of doubly charged gas ions are produced, resulting in a significant increase in the plasma density and plasma current [11]. However, the oscillating electrons also move toward the anode (chamber wall) due to the electric attraction from the anode [29–31] and some of them can escape from the HC. As d increases, the amount of the escaped electrons increases due to the weakening of the electrostatic confinement in the HC, resulting in the decrease in the number of ionization and excitation events and consequently, the decrease in the plasma density. In other words, as d increases from 8 to 24 mm, the nitrogen potential in the HC gradually decreases, resulting in a gradual decrease in the thickness of the compound layer (figure 10). When d decreases, the electrons emitted from the sample are decelerated in the sheath of the auxiliary cathode too quickly to induce full and efficient ionization and excitation of the gas. Consequently, a dark gap appeared in the HC, in which the intensity of the produced glow discharge plasma was even lower than that in the case without auxiliary cathode. Therefore, this is probably the reason that no significant nitridation layer can be formed at d = 6 mm.

The d-dependent variation of the nitrogen potential for nitriding correspondingly results in a change in the microhardness of the nitrided layer on the Cr12MoV alloy steel, as shown in figure 11. Under the condition of d = 8 mm, the nitrogen potential in the HC reaches the maximum level. In this case, the compound layer has the maximum thickness, while the nitrogen content in the diffusion zone needs be the maximal as well. Therefore, the nitrided layer has the maximum surface hardness of 1270 HV and the hardest diffusion zone. The hardness decreases with the increase in d as a result of a decrease in the nitrogen potential for nitriding. The hardness of the alloy steel nitrided at d = 6 mm is only slightly higher than that of the untreated sample. The result can be attributed to the fact that the nitrogen potential becomes extremely low due to the dark gap formation in the hollow cathode as discussed above.

Zoom In Zoom Out Reset image size