1.   Introduction

AISI 304 austenitic stainless steel has been selected in many applications because of its effective corrosion resistance, nonmagnetic property, and biocompatible performance. However, it has poor hardness, strength, and wear that should be improved [1–3]. As such, mechanical surface treatments (e.g., surface mechanical attrition treatment [4], shot peening [5–8], and ultrasonic nanocrystal surface modification [9]) and thermal surface treatments [10–11] (e.g., carburizing [12] and plasma nitriding [11,13–15]) have been performed to increase hardness and wear resistance.

Nitriding processes should be applied to create thick nitride layers and achieve a remarkable wear–fatigue resistance performance. This performance is attributed to the characteristics of a nitrided zone. Nitrided layer thickness varies depending on process type, operating temperature and duration, and chamber environment. However, high processing temperature and long duration cause distortion in metal parts, and this outcome is often undesirable. Plasma nitriding is performed under high-density plasma with low vacuum pressure; therefore, temperature and duration can be reduced without sacrificing the performance of nitrided surfaces [16]. This treatment has been improved through the effect of glow discharge in terms of voltage and current adjustments. Glow discharge provides ionization, and ions have been bombarded to surfaces via a high voltage difference [17–19]. Exposing plasma energy to thermally treated surfaces improves outcomes within a short time and at low temperature by accelerating diffusion; thus, a controllable microstructure is formed [20–23]. Zhao et al. [24] focused on enhanced plasma-assisted nitriding operations, such as plasma ion implantation, plasma immersion ion implantation, and thermo-ionically assisted triode. These treatments have been applied to industrialized parts regardless of shape and dimension [25].

Plasma nitriding has been widely selected to improve the surface of austenitic stainless steel. The industrial requirement for these steels is their corrosion feature. However, the performance of wear and fatigue resistance must be within a reasonable range. Therefore, various studies on different surface treatments have been conducted. Plasma (ion) nitriding is a thermo-mechanical operation and applied at a range of 350–550°C. Nitrogen atoms diffuse through the acceleration of plasma energy in a shorter time compared with that of conventional counterparts [26–27]. This process creates a body-centered tetragonal expanded austenite (S-phase) and consequently increases the surface hardness of austenitic stainless steels to HV 1000 [28–30].

For steels, an austenitic structure is prone to transformation into a metastable supersaturated nitrogen-rich S-phase at >450°C. The S-phase contributes high hardness and wear resistance performance. On the contrary, CrN precipitation begins to occur via the reduction of the Cr percentage in the matrix and leads to a remarkable loss of corrosion resistance performance [24]. Díaz-Guillén et al. [31] demonstrated the simultaneous combination of the S-phase and a chromium-nitrided structure via pulsed plasma nitriding within the range of 510–550°C. Lu et al. [32] showed that nitriding efficiency is low at 450°C. S-phase development with a high nitriding efficiency without degrading the corrosion resistance performance seems the main advantage of the treatment. Therefore, studies have focused on the rapid production of the S-phase [32].

This study aims to investigate the plasma nitriding preconditions in terms of phase alteration and transformation. Surface compound layer analysis, diffusion, and chemical changes were conducted to examine the fatigue–hardness performance. The previous literature results show that the treatment influences the S-phase, and the diffusion layer directly affects the microstructure and mechanical performance (wear, corrosion, and fatigue). Besides, various factors, such as frequency, gas mixture, furnace, process type (e.g., pulsed and active screen), and cycle, are considered for microstructural and mechanical alterations [18,23,30,33–35].

2.   Experimental

AISI 304 austenitic stainless steel specimens were manufactured with cold drawn bars after quenching and annealing at 1050°C. The specification, chemical composition, and mechanical properties of steel specimens are shown in the Table 1.

Table  1.  Specification, chemical composition, and mechanical properties of AISI 304 steel specimens

Specification
Dimension / nm	Shape			Tolerance				Length / m			Piece			Weight / kg
10000	Round			h9				3.0–3.2			566			1064
Chemical composition / wt%
C	Mn		Si		P		Cr		Ni		Cu	Mo			Co
0.04	1.68		0.4		0.03		18.24		8.0		0.6	0.26			0.17
Mechanical properties
Re (0.2%) / MPa		Rm / MPa				A / %				Z / %			Hardness (BHN)
341		748				39				70			261
Note: Re (0.2%)—Yield strength; Rm—Tensile strength; A—Total elongation; Z—Reduction in area.

下载: 导出CSV 

| 显示表格

The as-received and plasma-nitrided specimens with a diameter of ϕ25 mm and a thickness of 5 mm for microstructural characterizations are shown in Fig. 1. Rotating bending fatigue specimens were manufactured in accordance with the DIN 50113 standard.

Figure  1.  As-received and the plasma-nitrided with different conditions specimens for microstructural characterization analysis.

下载: 全尺寸图片 幻灯片

Plasma nitriding was performed in a PLC-controlled pulsed plasma nitriding furnace. A gas mixture composed of 25vol% N₂ and 75vol% H₂ was fed to the environment, and the ratio was chosen on the basis of the effectivity of 15vol%–30vol% of N₂ [30,36–37]. The gas mixture was then fed to a chamber at approximately 500 Pa under the potential of 500 V. The pressure within the range of 100–130 Pa was accepted [38]. Accordingly, 500 V and 333 Pa of potential and vacuum pressure, respectively, were selected for the novel approaches [31,39]. When the conditions of chambers were common for all nitriding processes, minimum preliminary parameters were determined on the basis of previously described approaches on the detection of the influence of nitriding temperature and duration on microstructural and mechanical performance. The effect of each parameter was determined by independently keeping the temperature and duration constant. The plasma nitriding conditions of the specimens are shown in Table 2.

Table  2.  Plasma nitriding conditions for specimens

Temperature / °C	Duration / h	Gas volume ratio / %
400	4	25 N2–75 H2
475	2	25 N2–75 H2
475	4	25 N2–75 H2
475	8	25 N2–75 H2
550	4	25 N2–75 H2

下载: 导出CSV 

| 显示表格

All the specimens were cut from cross-sections and grounded from 120-grade emery papers to 1200-grade emery papers and polished with 3, 1, and 0.25 µm diamond paste. Thereafter, the specimens were etched with a solution containing 100 mL of water, 100 mL of hydrochloric acid, and 10 mL of nitric acid for 180–210 s. The specimens were analyzed through optical microscopy (OM) and scanning electron microscopy (SEM) with Nikon Eclipse MA100 and Carl Zeiss Gemini Sigma, respectively. They were also examined through X-ray diffraction (XRD) by using a Rigaku Smart Lab diffractometer with Cu K_α radiation (scanning speed, 2°/min; voltage, 40 kV; scanning angle, 20°–90°; current, 30 mA). Micro-hardness tests were performed by using a Qness GmbH Q10 tester with a load of 0.005 N and a duration of 10 s. Fatigue tests were conducted with a BESMAK BMT-250 rotating bending fatigue tester by applying a frequency of 50 Hz and a constant stress amplitude of R = −1. Load was manually controlled with a load cell, and tests were performed via the DOLI software.

3.   Results and discussion

3.1   Characterization of the nitrided layer

The OM observations of the nitrided specimens are shown in Fig. 2. The nitrided layer can be distinguished via OM because of a strict boundary line between the core and the surface-treated layer. In general, the nitriding depth increases as process duration and temperature increase. The minimum nitriding depth is obtained at 400°C for 4 h, and microstructural alterations can be detected on the outmost layer. The thickness of the nitrided layers is similar at 475°C with the durations of 2 and 4 h. The nitriding depth remarkably increased at 475°C for 8 h and at 550°C for 4 h. The color of the layer also darkens under these conditions. Dark (black) precipitations begin to emerge at 475°C for 8 h, and the layer completely becomes black at 550°C for 4 h (Fig. 2).

Figure  2.  OM observations of plasma-nitrided specimens with different conditions: (a) 400°C for 4 h; (b) 475°C for 2 h; (c) 475°C for 4 h; (d) 475°C for 8 h; (e) 550°C for 4 h.

下载: 全尺寸图片 幻灯片

The SEM observations (Fig. 3) show that the nitrided layer was compatible with the OM. The nitrided layer thickness of the specimens treated at 475°C for 2 and 4 h seems similar under OM; however, the distinction was observed clearly through SEM analysis. Detailed analysis reveals the formation of dark precipitate at 475°C for 8 h, and the dark precipitate covers the whole nitrided layer at 550°C for 4 h. Energy dispersive X-ray (EDX) and XRD analyses supports the finding that chromium increases at 475°C for 8 h and at 550°C for 4 h (Figs. 4–5). Feugeas et al. [40] demonstrated that S-phase gradually and substantially transforms to CrN and Fe–N compounds through the application of plasma nitriding at high temperatures and long durations. Liang et al. [41] claimed that the S-phase may reach its maximum saturation level at 420°C.

Figure  3.  SEM observations of plasma-nitrided specimens with different conditions: (a) 400°C for 4 h; (b) 475°C for 2 h; (c) 475°C for 4 h; (d) 475°C for 8 h; (e) 550°C for 4 h.

下载: 全尺寸图片 幻灯片

Figure  4.  EDX elemental analysis of nitride layers at 475°C for 8 h.

下载: 全尺寸图片 幻灯片

Figure  5.  X-ray diffraction peaks of the as-received and nitrided specimens.

下载: 全尺寸图片 幻灯片

Shen et al.[1] stated that CrN forms on the nitrided layer at high temperatures, and Cr particles migrate from the grain interior to the boundaries. Shen et al. [29] applied plasma nitriding to AISI 304 in the range of 410–520°C for 4 h and changed vacuum pressure and potential within 410–420 Pa and 600–650 V, respectively. They achieved substantial CrN precipitation at 520°C. The duration was not adequate for CrN precipitation at 480°C, and SEM investigations and XRD results show that no precipitation occurs on the S-phase. The compound layer thickness observed at 410 and 480°C was consistent with that at 400 and 475°C for the same duration. Balusamy et al. [42] demonstrated the plasma nitriding of AISI 304 with direct current (DC) application at 500 V/2.67 Pa (potential/pressure) via 20%/80% N₂/H₂ volume ratio at 500°C for 8 h. Despite the high temperature and long duration, the compound layer was 4 µm. This result is attributed not only to processing temperature and duration but also to chamber potential, vacuum pressure, gas feeding performance, and other important treatment parameters.

XRD analysis (Fig. 5) reveals austenite peaks at different 2θ of the as-received specimens. Although a very thin influenced layer different from the core was observed at OM and SEM observations at 400°C for 4 h (Figs. 2(a) and 3(a)), no crystallographic alterations were detected in XRD analysis. Austenite transforms into an oversaturated nitride S-phase (γ_N) through plasma nitriding [43–44]. The conversion of austenite peaks to the S-phase exerts compressive residual stress on the nitrided layer [39]. Nitriding begins at 475°C for 2 h, and the S-phase and ɣ(111) peaks were observed. The S-phase was detected via low-temperature plasma nitriding, and its mechanical properties can be increased without losing the corrosion resistance performance. However, increasing the nitriding temperature and duration causes the conversion of S-phase to Fe₄N and CrN. Under different plasma nitriding conditions, CrN, Fe₄N, and Fe_2–3N forms [45]. Xia et al. [46] clarified the decomposition of S-phase and the occurrence of new chemical reactions of elements with nitrogen. Balusamy et al. [42] presented CrN precipitation at 500°C for 8 h and conducted XRD analysis; however, they could not observe precipitations through microstructural investigations. In some cases, the order of transformation can be changed. In the plasma nitriding of austenitic stainless steel (AISI 303), austenite is chemically converted to the S-phase, and the S-phase is transformed to ferrite and CrN precipitations. Solubility capacity increases as the operating temperature increases [47]. Thus, the decomposition of the S-phase leads to the formation of ferrite and CrN phases as revealed by the broadened ferrite peak and the raised CrN peak. Gontijo et al.[48] observed that nitriding at 500°C subsequently results in conversion to Fe₄N, Fe₂N, and α ferrite phases.

The changes in the microhardness of the nitrided specimens are presented in Fig. 6. The surface hardness reaches HV 1200 at 475°C for 8 h and at 550°C for 4 h. However, surface hardness can be increased up to HV 700 only when the temperature is 475°C and the durations are 2 and 4 h. Surface hardness remainsat HV 500 at 400°C for 4 h. Hardness improves because of the precipitation of CrN and the existence of γ_N phases [24]. CrN forms because of the excessive solution of nitrogen through the S-phase at high temperatures and long durations. CrN and Fe–N become stable and enhance hardness because of the S-phase [45]. The intensification of the compounds on XRD analysis is applied to surface hardness at 475°C for 8 h and at 550°C for 4 h.

Figure  6.  Change in the hardness of the as-received and nitrided specimens from the surface to the core. The SEM figures demonstrate the indentation trace shown by yellow lines.

下载: 全尺寸图片 幻灯片

The indentations on the nitrided zone are shown in Fig. 7. The interior indentation diagonals are two to three times longer than the surface indentation trace. Diffusion occurs from the surface to the interior during plasma nitriding. The nitrogen concentration profile abruptly decreases through the interior, and the effectivity of the S-phase, CrN, and Fe–N compound percentages gradually decreases. As such, the maximum level of hardness (the minimum diagonal trace) is accomplished on the outmost layer.

Figure  7.  SEM images of the nitrided specimen at 475°C for 8 h: (a) nitride layer; (b) indentation trace.

下载: 全尺寸图片 幻灯片

Two approaches are generally applied to determine the depth of a nitrided layer. In Model I, a boundary line was constructed by adding HV 100 to the core hardness; in Model II, a boundary line was created by increasing the hardness of 10% [49–50]. The nitrided and influenced depths based on the models were shown in Table 3.

Table  3.  Nitrided layer thickness based on models I and II

Plasma nitriding conditions	Nitrided layer thickness / µm	Influenced depth / µm
		Model I	Model II
400°C for 4 h	—	300	500
475°C for 2 h	3–4	450	950
475°C for 4 h	4–5	250	1000
475°C for 8 h	7–8	450	950
550°C for 4 h	30–32	650	1000

下载: 导出CSV 

| 显示表格

3.2   Fatigue resistance

The stress amplitude–number of cycle (S–N) curves of all the treated specimens are shown in Fig. 8. At a certain stress amplitude, the S–N curve of iron-based alloys flattens, and they can be assessed to have an infinite life. The fatigue limit of the as-received AISI 304 is 215 MPa. The limit increases to 260 MPa after nitriding at 400°C for 4 h. However, the layer is too thin to be measured at 400°C for 4 h. XRD analysis also revealed no transformation or occurrence on the nitriding layer. Conversely, hardness and fatigue limit slight increases. The reason for these changes can be expressed in terms of the high-temperature regime with nitrogen on the surface. For heating–cooling routes, the surface can be modified with a hard deposition. The fatigue limit increases to 355 and 600 MPa at 475°C for 2 h and at 550°C for 4 h, respectively. The endurance limits are close to each other at 475°C for 8 h and at 550°C for 4 h. Although the nitrided zone thicknesses are similar at 475°C for different durations, the effective fatigue resistance is enhanced during the treatment for 8 h. Hardness improvement and potential compressive residual stress application are the dominant mechanisms, not nitrided layer thickening. Moreover, the fatigue limit at 550°C for 4 h is the highest among the treatments because the diffusion layer thickens and hardens. It also retards crack occurrence and propagation because of the complete substantial transformation of the S-phase to a CrN/Fe–N compound layer, which is more stable and harder. Riazi et al. [51] demonstrated that plasma nitriding duration is more effective than temperature although compound layer saresix times thick. The results showed that the influence of preconditions (nitriding chamber and time–temperature conditions) should be concentrated and detected in terms of microstructural and mechanical behaviors.

Figure  8.  S–N curves of the as-received and plasma nitrided specimens.

下载: 全尺寸图片 幻灯片

Fig. 9 reveals the fatigue fracture surface of the as-received and treated specimens. In the as-received specimens, failure occurs through the formation of multi crack initiation sites supported by inner micro-cracks. At short nitriding durations, surface compression by residual stress and hard zone diminish the surface crack initiation zone, and subsurface inner micro-cracks has the dominant effect on the fracture (Fig. 9(c)). Inclusions are also one of the crucial mechanisms of fatigue fracture shown in Fig. 9(f). Inclusions create stress concentrations, so they behave as initial micro-crack formation points. Similarly, Wu et al. [52] proposed two fundamental approaches for fatigue fracture behaviors. In one of the approaches, a surface crack initiated and caused a fracture. In the other approach, a subsurface crack initially formed because of the prevention of a nitrided layer and various inclusion types, such as Al₂O₃, CaO, and other oxides, which were located on the subsurface.

Figure  9.  SEM observations of the fractured surface of different specimens: (a) as received; (b) nitrided at 400°C for 4 h; (c) nitrided at 475°C for 2 h; (d) nitrided at 475°C for 4 h; (e) nitrided at 475°C for 8 h; (f) nitrided at 550°C for 4 h.

下载: 全尺寸图片 幻灯片

4.   Conclusions

(1) Nitriding depth increased as process duration and temperature increase. The minimum nitriding depth was obtained at 400°C for 4 h. The nitriding depth remarkably increased at 475°C for 8 h and at 550°C for 4 h.

(2) An S-phase was observed when low-temperature plasma nitriding was applied and the hardness of the zone was increased. The S-phase was substantially converted to CrN precipitation at 475°C for 8 h and at 550°C for 4 h.

(3) The effective hardness on the surface improved under two nitride conditions (475°C for 8 h and 550°C for 4 h). The deepest nitrided zone was obtained at 550°C for 4 h.

(4) Plasma nitriding achieved surface hardness and increased fatigue limit regardless of process conditions; nevertheless, the best hardness and fatigue limit was obtained because of CrN precipitation.

(5) Fatigue fracture was observed in multi crack initiation sites in the as-received specimen. For the nitrided specimens, subsurface inner micro cracks and inclusions became the dominant effect on the fracture.

Acknowledgement

The authors would like to thank the Scientific and Technological Research Council of Turkey (TUBITAK) for the support of this study (Grant No: 215M134).