Regular Article
Improvement of the Tribological Properties of DLC/oxynitriding Duplex-treated AISI H13 Alloy Steel
2014 Volume 54 Issue 1 Pages 193-198
Details
2014 Volume 54 Issue 1 Pages 193-198
In this study, diamond-like carbon (DLC) films were prepared by DC-pulsed plasma CVD after the oxynitride treatment of AISI H13 tool steel. In order to investigate the tribological properties of DLC films, a Raman spectroscopy analysis, wear test, adhesion and roughness tests were performed. The main parameters of the DC-pulsed plasma CVD process includes various pretreatment times of argon plasma (15, 30, 45 and 60 min). Experimental results showed that an oxynitride layer and the DLC films could be completely obtained after DLC/oxynitriding duplex treatment. The duplex coating layers had optimal adhesion (critical load reached to 10.65 N) and wear properties after DC-pulsed plasma was CVD treated via a low pulse voltage (–1.5 kV), pretreatment times of the argon plasma were 15 min and the substrate temperature was kept at 40°C. Meanwhile, the optimal DLC/oxynitriding duplex treated specimens possessed the lowest wear volume (2.25 × 10–3 mm3) and a lower friction coefficient (0.06).
AISI H13 alloy steel is a high strength steel with common uses in industry as a raw material for both hot and cold applications.1) Furthermore, the H13 alloy steel is normally used for plastic and die-casting dies, given its good temperature resistance, high hot hardness strength, and good resistance to thermal fatigue and wear.2,3) Our previous studies mostly aimed at enhancing the surface performance of the H13 steel, and thus, focused on the development of single coatings and/or thermo-chemical treatments to protect the die surface.2,3,4)
To extend the die life, nitriding treatment is often applied to the die surface. Nitriding is a thermo-chemical treatment with nitrogen diffusion, which leads to the surface hardness of the treated dies, due to compressive internal stress within the nitrogen-enriched layer. It shows an effectively increase of thermal fatigue characteristics due to the effects of compressive stress and the hardened surface.2,4) During the oxidation process, most steels can form several kinds of oxides (Fe2O3 and Fe3O4). Generally, the Fe3O4 layer of the oxidation treatment can effectively protect and improve the erosion and corrosion resistance of steels. Our previous study found that oxynitriding specimens could form a passive film, which contributed a better property during the duplex coating treatments.1,4,5)
Diamond-like carbon (DLC) films have excellent properties such as high hardness, low friction, and chemical inertness.6,7,8) Furthermore, good adhesion to steel and low price preparation is desired to avail the DLC film coating more widely utilize.6) In addition, there are many previous experimental results suggest that plasma CVD using pulsed DC discharge could be effective in improving the adhesion, where specimens on the substrate can be controlled by the conditions of the pulse bias.6,7,8,9)
Because of AISI H13 alloy steels are almost exclusively used on cold and hot tooling applications, thermal fatigue and wear resistance are the most important factors in limiting tool life and resulting in the tool failure of H13 tool steels.2,3) Surface treatment is usually used to increase wear resistance as well as to improve surface hardness and tool life.10) In this study, the oxynitriding process uses steam at the end of the nitriding stage as an oxidizing medium, which is an integral part of the treatment. Furthermore, DLC has many superior properties, such as high mechanical hardness, high wear resistance, a low friction coefficient and chemical inertness. Therefore, the DLC/oxynitriding duplex treatment was utilized for treating AISI H13 tool steel in an attempt to increase tool life.
In this study, AISI H13 alloy steel was chosen as the substrate material to undergo a homogeneous heat treatment (quenched at 1030°C and tempered at 580°C for 3 h, repeated 3 times to reach a hardness of 47±1 HRC). Therefore, a typical microstructure was obtained through commercial heat treatment, which comprises the structure of tempered martensite and pro-eutectoid carbides. The chemical compositions (mass%) of AISI H13 alloy steel are as follows: 0.34% C, 0.81% Si, 0.38% Mn, 0.14% Ni, 5.0% Cr, 1.15% Mo, 0.67% V and 91.4% Fe. The oxynitriding specimens were nitrided for 1 h at 530°C and oxidized via steam for 30 min at 525°C.
The DLC/oxynitriding duplex treatments, which underwent DLC coating with oxynitriding specimens, were treated the same as in the above mentioned process. DLC coated oxynitriding specimens used the DC-pulsed plasma CVD technology, as shown in Fig. 1. After the chamber was pre-evacuated using a rotary pump, a CH4 mixture of gas was supplied to the chamber. To study the effects of the different pretreatment times of argon plasma on the DC-pulsed plasma CVD process, the different pretreatment times included 15, 30, 45 and 60 min. Meanwhile, the duty cycle of the DC-pulsed plasma CVD was maintained at 10%; the pulsed voltage and frequency were kept at –1.5 kV and 1 kHz, respectively. In this study, due to the limitations of our laboratory equipment; if the frequency of DC-pulsed plasma CVD increased to 10 kHz, it easily generated the charge accumulation and a glow-discharge phenomenon. This would not help produce DLC coating; therefore, the research selected a low frequency of 1 kHz. Moreover, CH4 gas (5 sccm) was added at less than 10–2 torr and continued for 2 h, sing temperature to 500 followed by depositing of the DLC films. The experimental results showed that a 50–60 μm oxynitride layer and 1–2 μm of DLC film could be obtained after the AISI H13 alloy steel was treated by DLC/oxynitride duplex treatment.
Schematic of the DC-pulsed plasma CVD system.
The aim of this study is to investigate the properties of DLC films using different pretreatment times of argon plasma for the DC-pulsed plasma CVD treatments. In order to evaluate the properties of DLC films and tribological behaviors for DLC/oxynitriding duplex-treated AISI H13 alloy steel, a Raman spectroscopy analysis (MOF-iHR550), wear test (POD-FM406), roughness test (ET4000A), scratch tests (Scratch Tester J & L Tech. Co., Korea) and SEM (Hitachi-S4700) microstructure inspections were performed. The wear resistance of the specimens was evaluated in a ball-on-disk test (ASTM G99). The wear test parameters were as follows: the specimen size was ϕ36 × D5 mm, diameter of WC ball (HRA 90±1) was 6 mm, axial load was 2 N, disk rotation was 200 rpm, sliding speed was 0.63 ms–1 and total rotation was 10000 revolutions.
Figure 2(a) shows Raman Spectroscopy for the non-treated and different pretreatment times of argon plasma. By using Gaussian function dismantling and synthesis to calculate the results of integration area ratio (ID/IG) and the offset of G-peak, as shown in Fig. 2(b), this study compared it with the non-treated specimen (ID/IG was 0.70), and found that the ID/IG value (0.71) showed insignificant diversification after argon plasma pretreatment for 15 min. However, by increasing the time of argon ion bombardment (30 min), the ID/IG values (1.25) rose rapidly. Moreover, this shows a slow decline as the argon plasma bombardment increased to 45 min, and then, maintained a flat trend (60 min). The results are similar to the offset of G-peak. When the pretreatment time for argon ion bombardment was 15 min, the G-peak shifted slightly; moreover, it rapidly rose at pretreatment times for argon for 30 min, and then maintained a flat trend after 45 min. The lowest G-peak (1544 cm–1) for the different pretreatment times for argon plasma appeared in 15 min. According to the above results, this result should be related to the character of DLC films. Increasing the pretreatment times of argon ion bombardment will results in an increase in the substrate temperature. The results show that the DLC films gradually approaching graphitization.
Comparison of the (a) Raman spectroscopy, (b) ID/IG and G-peak of Raman analysis by non-treated and different pretreatment times of argon plasma.
Figure 3 shows the substrate temperature after different pretreatment times for argon plasma. The substrate temperature was slowly enhanced as the deposition times increased after pretreatment of argon plasma for 15 min; nevertheless, the substrate temperature was still less than 40°C. Conversely, the substrate temperature seemed to decline slightly as the deposition times increased after pretreatment of argon plasma for 30, 45 and 60 min. However, the substrate is almost maintained at a higher temperature (50 to 55°C). Budtz-Jørgensen et al.11) indicated that the substrate surface temperature of the deposited film was related to the ion kinetic energy. Therefore, the substrate temperature was higher after the argon plasma bombardment for 30, 45 and 60 min. It is reasonable to suggest that the higher energy easily forms a graphite structure (sp2 bonds). Conversely, due to a much lower ion kinetic energy after argon plasma bombardment for 15 min, the substrate temperature was not raised rapidly; causing the DLC films to form a more stabilized bonding (sp3 bonds). This result agrees with our previous discussions of Fig. 2.
Comparison of the substrate temperature by different pretreatment times of argon plasma.
Significantly, as the pretreatment time for argon plasma increases (15 → 30 → 45 → 60 min), the ID/IG value gradually increases, and then tends to slow down (as shown in Fig. 2). According to the above discussion and results, it is reasonable to suggest that the pretreatment time for argon plasma of 15 min should be the optimal process for DC-pulsed plasma CVD. Figure 4 shows the surface morphology observations of the scratch test after the different pretreatment times for argon plasma. It was evident that the DLC coating films were peeled off after a much longer distance (7.5 mm) by the argon plasma pretreatment for 15 min, as shown in Fig. 3(a). Compared with other argon plasma pretreatments (30, 45 or 60 min), the other DLC coating films peeled off at the beginning of the scratch test, as shown in Figs. 4(b)–4(d). Further analysis of the appearance of specimens after the scratch test, it was confirmed that the argon plasma pretreatment for 15 min yields better DLC coating adhesion.
Surface morphology observations of scratch test by different pretreatment times of argon plasma (a) 15 min, (b) 30 min, (c) 45 min, (d) 60 min.
Figure 5 shows a slight variation in the friction coefficient of the scratch test after different pretreatment times of argon plasma. Compared with the other pretreatments, the friction coefficient of argon plasma pretreatment for 15 min is relatively stable and lower. When increasing the pretreatment times of argon plasma (30 → 45 → 60 min), the friction coefficient seems to have a gradual rising trend after a scratch distance of 1 mm. Significantly, the critical load of DLC coating films reached 10.65 N, which has better adhesion with argon plasma pretreatment for 15 min.
Comparison of the variation of the friction coefficient of scratch test by different pretreatment times of argon plasma.
Figure 6 shows the friction coefficient of the wear test after different pretreatment times of argon plasma for DC-pulsed plasma CVD treatments. The lowest coefficient of friction (0.06) appeared in argon plasma pretreated for 15 min. Furthermore, the highest coefficient of friction (0.27) appeared in argon plasma pretreated for 30 min. Suzuki et al.12) noted that with the increase in pretreatment times for argon plasma, the surface roughness and substrate temperature increased. The surface roughness of non-treated argon plasma specimens was 0.20 μm, and only increased a little (0.23 μm) after argon plasma pretreatment for 15 min, as shown in Fig. 7(a).
Comparison of the friction coefficient of wear test by different pretreatment times of argon plasma.
Comparison of the surface roughness and volume loss of wear test by different pretreatment times of argon plasma.
In addition, the argon ion bombardment contributes to a good clean specimen surface; but if the substrate temperature rises too much it will result in unstable bonding of the DLC films (sp3 → sp2). As a result, the 15 minutes argon plasma pretreatment has a slightly higher surface roughness and the lowest coefficient of friction, and the pretreatment times of argon plasma (30 → 45 → 60 min) possess a higher coefficient of friction (as shown in Fig. 6), which could be related to unstable bonding of DLC films. However, increasing the argon plasma pretreatment times can provide a higher energy to generate ion bombardment. It can also increase the surface roughness of the specimens. The highest surface roughness appeared in argon plasma pretreated for 45 min, and declined at 60 min (Fig. 7(a)). According to the literature13) indicated that continued to increase the argon plasma pretreatment times, the surface roughness will decline. In general, the roughness of DLC films has been found to depend on the ion energy, substrate temperature, substrate materials and film composition. If the plasma bombardment time is longer (60 min), which can cause the sp2 bonding contents decrease. Meanwhile, the sp3 bonding contents and the compressive stress are increased. Therefore, it will be obtained a smoother surface and lower friction coefficient of DLC films that result in a decline of roughness.
Figure 7(b) shows the volume loss after a wear test with different pretreatment times for argon plasma. The lowest (2.25 × 10–3 mm3) and highest (3.41 × 10–3 mm3) volume loss appeared in argon plasma pretreated for 15 and 30 min, respectively. In addition, the tribologists usually use relative wear rate (mm3/Nm) instead of wear volume (mm3), because the relative wear rate is independent of conditions in ball-on-disk test. In this study, the lowest and highest wear volume of 2.25 × 10–3 and 3.41 × 10–3 mm3 will be correspond to the relative wear rate of 5.95 × 10–7 and 9.02 × 10–7 mm3/Nm respectively. This value is slightly lower than relative wear rate of conventional DLC films deposited by plasma CVD. However, the experimental conditions were restricted since our test equipment did. It was reasonable to suggest that the lower relative wear rate resulted from the low substrate temperature (less than 40°C) during deposition and low pulse frequency (1 kHz). The results can be further compared with Fig. 6. The volume loss of the specimens agrees with a variation in the coefficient of friction. Previous studies indicate14) that the deep wear tracks on the films are mainly due to plastic deformation caused by rubbing under high contact-pressures, and not due to wear loss. According to the above results and discussion, the optimal parameter of pretreatment times of DC-pulsed plasma CVD was 15 min.
Furthermore, the surface morphology after wear tests was observed, as shown in Fig. 8. Non-treated specimens and argon plasma treated for 15 min possessed a more smooth appearance, as shown in Figs. 8(a) and 8(b), and the latter specimen, especially, had a relatively flat morphology. It was also confirmed that the minimum volume loss from the wear test was for the argon plasma specimens treated for 15 min (as seen in Fig. 7(b)). Moreover, after increasing the pretreatment times of argon plasma, significant wear appeared on the surface morphology of DLC coating, as shown in Figs. 8(c), 8(d) and 8(e). It is reasonable to suggest that significant wear is related to the plastic deformation, which is disadvantageous to the variation in the coefficient of friction (as shown in Fig. 5). According to the above discussion and results, argon plasma pretreatment underwent the DC-pulsed plasma CVD process that played an important factor in affecting the bonding of DLC films. The failure-resistant capability of DLC films created an improvement after the optimal argon plasma pretreatment (15 min).
Surface morphology observations of wear test by non-treated and different pretreatment times of argon plasma (a) non-treated, (b) 15 min, (c) 30 min, (d) 45 min, (e) 60 min.
Further analysis was done on the elements of wear for untreated specimens and those with different argon plasma pretreatments, as shown in Table 1. Due to the DLC films being prepared by DC-pulsed plasma CVD after the oxynitride treatment of AISI H13 tool steel, the EDS results indicated that most of the elements on the wear surface morphology were Fe, O, C and Cr; W elements appeared in very small amounts. It can be confirmed that the appearance of wear specimens still exists in the DLC and oxynitride layers. The above results also agree with the Raman spectroscopic analysis of the ID/IG values (Fig. 2(b)): the lower ID/IG values usually possess a better stable bonding and emerge with better wear resistance.
| Position | Elements (at.%) | ||||
|---|---|---|---|---|---|
| Fe | Cr | O | W | C | |
| a-1 | 34.58 | 0.53 | 59.93 | – | 4.95 |
| a-2 | 36.28 | 0.34 | 63.38 | – | – |
| a-3 | 13.04 | 0.17 | 24.45 | – | 62.53 |
| b-1 | 12.81 | 0.16 | 22.61 | 0.61 | 63.80 |
| b-2 | 8.48 | 0.12 | 12.55 | – | 78.86 |
| b-3 | 36.52 | 0.67 | 62.81 | – | – |
| c-1 | 34.75 | 0.53 | 59.13 | – | 5.59 |
| c-2 | 15.86 | 0.23 | 29.12 | – | 54.79 |
| c-3 | 17.03 | 0.31 | 42.68 | 1.16 | 38.82 |
| d-1 | 35.69 | 0.85 | 63.46 | – | – |
| d-2 | 10.60 | – | 33.32 | 0.51 | 55.56 |
| d-3 | 14.29 | 0.19 | 25.02 | – | 60.51 |
| e-1 | 35.69 | 0.85 | 63.46 | – | – |
| e-2 | 35.62 | 0.88 | 63.5 | – | – |
| e-3 | 14.29 | 0.19 | 25.02 | – | 60.51 |
Due to the lower ID/IG value (0.71) and maximum offset of G-peak (1544 cm–1) appearing in the DLC/oxynitriding duplex-treated AISI H13 alloys steel after argon plasma pretreatment for 15 min, the study results show that this coating layer possesses optimal adhesion, a stable microstructure and better wear resistance. Consequently, the optimal parameter of pretreatment times for argon plasma for DLC/oxynitriding duplex-treated AISI H13 alloys steel was 15 min. The optimal process of DC-pulsed plasma CVD obviously improved the tribological properties of DLC/oxynitriding duplex-treated AISI H13 alloy steel.
An oxynitride layer and DLC films can be obtained after DLC/oxynitriding duplex treatments. The optimal parameters of pretreatment times of DC-pulsed plasma CVD was 15 min. It possessed a relatively low substrate temperature (35°C); meanwhile, it can form stable bonding (sp3 bonds) and attain optimal adhesion of DLC films (critical load reached 10.65 N). Moreover, the lower ID/IG value (0.71), the greater the maximum offset of G-peak (1544 cm–1) and the better the friction coefficient (0.06), the lower the wear volume (2.25 × 10–3 mm3) obtained. Consequently, DLC films are effective in improving the tribological properties of DLC/oxynitriding duplex-treated AISI H13 alloy steel by using the argon plasma pretreatment of DC-pulsed plasma CVD technology for 15 min.
This research is supported by the ASSAB STEELS TAIWAN CO., LTD. The authors would like to express their appreciation for Professor C. M. Liu.
