Mechanical Properties
Effect of Selenium on the Machinability of As-hot Rolled and Heat Treated 4140 Steel
2020 Volume 60 Issue 4 Pages 782-791
Details
2020 Volume 60 Issue 4 Pages 782-791
This study aims to clarify the effect of selenium (Se) on the machinability of AISI 4140 steel in the as-hot rolled (HR) and quenched and tempered (QT) conditions. Machining tests were conducted to examine the progressive tool flank wear and the chip formation. The characteristics of the workpiece materials such as hardness, microstructure, and non-metallic inclusions were investigated. The worn tools were analyzed to characterize the possible deposit formed on the tool surface. In addition, micro-computed tomography (µ-CT) and high-resolution transmission electron microscopy (TEM) were utilized to study the Mn(S,Se) inclusions in severely deformed chips.
The investigation showed that Se micro-alloying did not improve the machinability of the HR 4140 steel. However, in the QT condition, the superior machinability was observed for the Se-treated 4140 steel when compared with the untreated steel. Therefore, the effect of Se on improving steel machinability was considered to be strongly dependent on the structure and properties of metal matrix. The factors influencing the improvement in machinability of Se-treated QT steel are discussed.
With advances in steelmaking technologies, the concentration of impurities in steel such as oxygen, phosphorus, hydrogen, and nitrogen have been reduced to extremely low levels.1,2) Clean steel grades generally have excellent mechanical properties (fatigue strength, ductility, and toughness) and corrosion resistance. However, these high-performance steels also create production problems during part machining. For example, more difficult chip breaking, severe tool wear, and an increase in manufacturing cost are often observed. Therefore, the production of high-performance steels with improved machinability is an important focus for the future steelmaking technology. Machinability is defined as the ease with which a material can be machined by a given tool, which can be represented by a number of parameters, such as chip formation and fracture, tool wear, tool life, surface integrity of the machined material, and the power consumption during machining.3,4) Therefore, the machinability of steel is dependent on the characteristics of the workpiece, the cutting tool, and the machining parameters.5,6) Chinchanikar et al.7) investigated the machinability of AISI 4340 steel with different levels of hardness. For the harder workpiece, higher cutting force but a better surface finish was observed. Das et al.8) studied the influence of machining parameters (cutting speed, feed rate, and depth of cut) on the machinability of AISI 4140 steel. The surface roughness of the machined steel was found to be mainly influenced by feed rate and cutting speed, while the tool flank wear was primarily affected by cutting speed and the interaction of feed rate-depth of cut.
To improve the machinability, the addition of free-cutting elements, such as lead, sulfur, and selenium to steel has been suggested.9) Historically, lead had been added to many steels to improve the surface finish and reduce tool wear; however, lead is potentially harmful and alternative lead-free steels have been developed more recently.10,11) Ånmark et al.12,13) investigated the machinability of different carburizing steel grades alloyed with sulfur. It was found that a high content of MnS inclusions, large grain size, and low hardness were beneficial for achieving superior machinability. Jeon et al.14) studied the effect of sulfur content on the machinability of 27Cr-7Ni duplex stainless steel. With increasing sulfur content from 0.005% to 0.1181%, the tool life was increased 2 to 12 times. Increasing sulfur content in re-sulfurized steel can improve the machinability, but it can also degrade the mechanical properties of steel, such as toughness, ductility, and pitting corrosion resistance.15)
Selenium and tellurium belong to the same group of elements in a periodic table as sulfur, and even a small amount of these elements can improve the machinability of carbon steels. However, the mechanism of such improvement in machinability by micro-alloying with these elements is still under investigation16,17,18,19,20) and is not clear. Gol’dshtein et al.16,17,18) investigated the effect of Se on the properties of structural steel. The investigators found that the addition of Se to steel promoted the formation of globular sulfides and an improvement to machinability was confirmed. It is interesting to note that the ductility and toughness of the steel were also improved by small Se additions in this study. Zaslavskii et al.19,20) studied the influence of Se on the machinability of maraging steels. An improvement in the machinability due to Se micro-alloying was observed. The mechanism was considered to be related to the formation of a soft selenide film between the machining tool and the chips which reduced the friction between the workpiece and the tool and thus increased the tool life.
Although basic studies of the effect of Se on steel machinability have already been performed over the last several decades, modern steelmaking practice requires optimization of the micro-alloying treatment to improve machinability and mechanical properties. Moreover, to the authors’ best knowledge, no investigations on the effect of Se on the machinability of steel with different metal matrix microstructures have been reported, which is important for industry to reduce manufacturing cost and increase labor productivity. Therefore, the current work aims to clarify the effect of Se on the machinability of 4140 steel with two different properties and microstructures: as-hot rolled (HR) and quenched and tempered (QT) 4140 steels.
Four samples of industrially produced AISI 4140 steels were investigated in this study. Table 1 shows the chemical composition of the workpiece materials. The first test set was in the HR condition: steel #1A was Se-treated and steel #1B was in the base untreated condition. The second test set included QT steels: Se-treated steel #2A and base untreated steel #2B. Induction heating was used for the QT heat treatment and the details are as following: austenitizing at 890°C for 0.5 ks, quenching by water at 38°C for 0.1 ks and tempering at 685°C for 0.5 ks. All of the 4140 type steels studied were supplied as bars of approximately 95 mm in diameter and 300 mm in length. Because industrially produced 4140 steels from different heats were used in this study, there were minor variations in alloying elements.
| Name | Steel | Status | C | Si | Mn | S | Cr | Mo | Ni | Al | Ca | Cu | Se |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| #1A | 4140Se | HR | 0.42 | 0.29 | 0.94 | 0.030 | 1.05 | 0.24 | 0.21 | 0.008 | 0.0016 | 0.2 | 0.0173 |
| #1B | 4140 | HR | 0.42 | 0.16 | 0.91 | 0.026 | 0.95 | 0.16 | 0.09 | 0.024 | 0.0013 | 0.18 | – |
| #2A | 4140Se | QT | 0.43 | 0.22 | 0.94 | 0.033 | 1.04 | 0.27 | 0.21 | 0.008 | 0.0012 | 0.24 | 0.0209 |
| #2B | 4140 | QT | 0.41 | 0.23 | 0.94 | 0.024 | 1.06 | 0.16 | 0.11 | 0.027 | 0.0015 | 0.21 | – |
Figure 1 shows the experimental equipment. In this investigation, single point turning tests were performed with lubricant (6–7% emulsion) on a HAAS TL-1 CNC lathe (Fig. 1(a)). Several Sandvik Coromant cemented carbide cutting tools with chip breaker (SNMG 431-QM 4325) were utilized for the machining tests (Fig. 1(b)). Machining parameters were: 80 m/min cutting speed, 0.204 mm/rev feed rate, and 1.275 mm depth of cut. In each test, fifteen passes were machined on each bar and several bars were used to get a critical tool flank wear (about 0.15 mm). After machining each bar, the progressive tool flank wear was accurately measured using a scanning electron microscope equipped with backscattered electron detector (SEM-BSE). For example, the SEM image of the tool flank wear of the steel #2B was shown in Fig. 1(c). The chips formed during machining were also collected for analysis.
CNC lathe machining test (a), cutting tool (b), and SEM image of tool flank wear (c). (Online version in color.)
Hardness profiles were obtained on the transverse section of the tested steel bars using the Brinell method with 3000 kg load and a 10 mm tungsten carbide ball. The specimens for material characterization were cut from the half radius position of each steel bar, then ground using sand paper and polished to 0.1 μm using a diamond suspension. The morphology, size, and compositions of the non-metallic inclusions in the workpiece materials were analyzed in the longitudinal direction by an automated scanning electron microscope equipped with energy dispersive spectroscopy (SEM/EDX). A joint ternary diagram analysis was used to characterize the non-metallic inclusions in each sample. Each point on this diagram represents one inclusion and the three dominant elements in overall inclusion composition, and the markers indicate the average diameter of inclusion. The microstructure and micro-hardness of the specimens were also characterized using an optical microscope (OM) and a Struers Duranmin Hardness Tester, respectively, after etching with a 2% nital solution. Furthermore, the deposits on the rake surfaces of the worn tools formed during machining were also characterized by SEM/EDX analysis.
Several special methods were employed to identify the possible influencing mechanisms of Se micro-alloying on machinability:
- Electrolytic extraction in a 10% acetyleacetone electrolyte solution with 40–60 mA of applied electric current was used to extract non-metallic inclusions from workpiece and machining chips. The extracted inclusions were collected on a polycarbonate filter with an open pore size of 0.2 μm and characterized using automated SEM/EDS analysis;
- A high resolution μCT scan (ZEISS Xradia 510 Versa) was used to analyze the 3D spatial distribution of non-metallic inclusions in a machining chip. The machining chip with approximately 6 mm × 3 mm × 0.1 mm dimensions was used directly for X-ray CT scanning. Totally 1601 projections with a resolution of 1 μm/pixel were obtained.
- A high resolution transmission electron microscopy (TEM) analysis was performed on a specimen extracted from a machining chip at an accelerating voltage of 200 kV. The TEM specimen was prepared using a focused ion beam (FIB) lift-out technique in a FEI Helios NanoLab 600.
The hardness profiles of the workpiece materials through the cross section of bars are shown in Fig. 2. Regarding the HR 4140 steels, the Se-treated steel #1A showed slightly higher hardness (approximately 290 HB) than the base steel #1B (about 275 HB), which may be from the combined effects of a higher alloy content (Cr, Mo and Ni)21) and the Se addition in steel #1A. In the second set of QT steels, steel #2A (4140 Se) and #2B (4140 base) had almost the same hardness values (320 HB to 330 HB) in the position of bars that were machined, as shown in Fig. 2 by the arrow.
Hardness profiles of the workpiece steels on different locations. (Online version in color.)
The microstructural features of the workpiece materials indicate that the HR base #1B and Se-treated #1A had a uniform distribution of ferrite and bainite (Figs. 3(a) and 3(b)). Microstructure of the QT base #2B and Se-treated #2A consisted of fine tempered martensite which was formed during the quenching and tempering treatment (Figs. 3(c) and 3(d)).
Microstructures of the 4140 steel bars (a, b, c and d) and chips (a’, b’, c’ and d’). (Online version in color.)
Figure 4 shows the SEM-BSE images of typical non-metallic inclusions found in the longitudinal sections of these steels. Three kinds of inclusions were found in the base HR steel #1B: deformed pure MnS, Al–Ca oxide, and complex MnS with Al–Ca oxide core (see Figs. 4(d)–4(f)). In HR steel #1A, the Se addition promoted the formation of complex alloyed Mn(S,Se) inclusions, which were present as a mixture of elongated and nearly spherical shapes with different Se contents (Figs. 4(a)–4(c)). It appears that the morphology of these inclusions are dependent on the ratio between Se and S in the inclusion: MnS inclusions that were highly alloyed by Se were visible in cross section and had a more round shape. Similarly, the QT Se-treated steel #2A also contained mostly complex alloyed Mn(S,Se) inclusions and a few small complex Al–Ca based oxy-sulfides (see Figs. 4(g)–4(i)). In base QT steel #2B, the inclusions consisted of elongated MnS and some Al–Mg oxide wrapped by MnS (see Figs. 4(j)–4(l)).
Typical inclusions in steels: #1A (a–c), #1B (d–f), #2A (g–i), and #2B (j–l).
Approximately 1000 non-metallic inclusions were counted in each specimen using automated SEM/EDX analysis. The total area analyzed depended on inclusion density per unit of area (1 mm2) in each sample. The scan was terminated once the targeted number of inclusions was reached. The families of non-metallic inclusions in these steels are presented in joint ternary diagrams (Fig. 5), where the color indicates the average diameter of inclusion.
Joint ternary diagrams of non-metallic inclusions in the steels: 1A (a), 1B (b), 2A (c), and 2B (d). (Online version in color.)
The total concentrations of elements within the inclusions, in ppm, characterizes the chemistry associated with the inclusions in the matrix, are shown in Fig. 6(a). It was observed that Mn and S are the major elements which formed inclusions in all of the steels studied. Se was detected in both of the Se-treated steels (#1A and #2A). The frequency distributions for equivalent diameter and aspect ratio of inclusions in the QT steels are shown in Figs. 6(b) and 6(c), respectively. The non-metallic inclusions in these workpiece materials were mainly distributed in the range of 1 to 10 μm of equivalent diameter, but the Se-treated steel #2A had a larger number of small inclusions than the base steel #2B.
Total concentrations of inclusions (a), and frequency curves of the average diameter (b) and aspect ratio (c) of inclusions in the QT steels. (Online version in color.)
To understand the effect of Se micro-alloying on the machinability of 4140 steel with different metal matrix properties and microstructures (HR vs QT), four steels (Table 1) were machined and the tool wear was periodically measured after machining each bar. The change in tool flank wear during machining of each of the 4140 steels is presented in Fig. 7. For the base 4140 steels, the tool wear rate was significantly higher in the QT condition (#2B) when compared to the HR condition (#1B). The large difference in tool wear rate for the two base steels with same chemistry grade is likely due to the higher hardness in the QT condition (Fig. 2). Therefore, machining high strength steel in QT condition is often problematic in industry. Micro-alloying of the QT steel #2A with Se decreased the tool wear rate by about three times and brought it to the level that observed in HR condition. However, Se micro-alloying did not improve the machinability of the HR 4140 steel with a softer metal matrix when compared to the QT condition. In fact, the machinability of the HR 4140 steel was slightly degraded by Se micro-alloying. These tests showed that the effect of Se micro-alloying on the machinability of 4140 steel is strongly dependent on the properties and microstructure of the workpiece material.
Tool flank wear during machining of the base and Se-treated 4140 steels with different status. (Online version in color.)
Figure 3 compares the microstructures of workpieces and collected chips. Se micro-alloying did not have an observable effect on the microstructure of bars. However, heavily deformed microstructural features in the machining chips were observed during machining of the softer microstructure in the HR steel when compared to harder QT steel for both the base and Se-treated steels. The chips showed an increased micro-hardness compared to workpiece and more work hardening was observed during chip forming in HR steel (see Fig. 8) compared to the QT steel.
Comparison of the micro-hardness of steel bars and chips.
Figure 9 illustrates the morphologies of typical inclusions in the chips after machining of the base and Se-treated QT steels. The red arrows in the figure show the directions of chip deformation. It was found that both the pure MnS and the complex Mn(S,Se) inclusions in two types of steels were severely deformed along the deformation direction of chips during machining. Sulfides were often detected near a crack tip, indicating that these inclusions promoted chip breaking. More detailed 3D μCT observation (Fig. 9(c)) clearly showed the directional orientation of Mn(S,Se) inclusions in severely deformed chips of Se-treated steel and also indications of crack initiation by the inclusions.
Observation of sulfide inclusions in chips after machining QT 4140 steels: (a–b) polished and (c) μCT of #2A Se, and (d–e) polished #2B base. (Online version in color.)
In addition to μCT scanning, an electrolytic extraction method was also used to observe the real 3D shape and chemistry of non-metallic inclusions in workpiece and machined chips of QT steels. The insert in Fig. 10(a) illustrates the three-dimension morphology of the extracted Mn(S,Se) inclusions in chips of steel #2A. The aspect ratio of these inclusions was calculated as a ratio of max/min diameters measured by the automated SEM/EDX method. Figure 10(a) also shows the aspect ratio distribution of the inclusions. It was found that more inclusions with a large aspect ratio were observed in chips compared to that in steel bar, which indicated that the inclusions deformed during machining. Moreover, the chemistry of the extracted inclusions was correlated to their aspect ratio (Fig. 10(b)). The statistical analysis showed that the Se-rich inclusions had a lower aspect ratio, which indicates that Se micro-alloying promotes sulfide shape modification.
Frequency curve of aspect ratio of 3D extracted inclusion (a) and effect of Se/S chemistry ratio on aspect ratio of inclusion (b) in chips of the machined QT Se-treated #2A steel.
To better understand the effect of inclusions, especially Mn(S,Se) inclusions on chip formation and steel machinability, a FIB-TEM investigation with the extraction of a sample from the machined chip of Se-treated QT steel #2A was performed (Fig. 11). Two Mn(S,Se) inclusions (zones B and D) with the surrounding matrix (zones A and C) were extracted and investigated. It was confirmed by a selective area electron diffraction that Mn(S,Se) inclusion formed in the Se-treated 4140 steel was a solid solution sulfide (Fig. 11(b)). The lattice parameter of Mn(S,Se) was determined to be 5.19 Å, which was about 5% larger than that of a pure MnS (4.95 Å). Considering that atomic radius of Se is 15% larger than the atomic radius of S, the presence of 20–30% Se in MnS is expected to increase the lattice parameter of the complex Mn(S,Se) up to the 5%. That was observed. This observation indicates that solid solution strengthening of MnS by the addition of Se occurs. As a result, the machinability of the HR 4140 steel (steel #1A) was slightly degraded after alloying by Se compared to that of the base HR 4140 steel (steel #1B), as seen in Fig. 7. In addition, it was observed that the inclusion deformed significantly along <100> direction family during machining and the lattice distortion is ± 5.5 degrees. The observation of the inclusion/matrix interface showed an interfacial crack between the Mn(S,Se) inclusion and the steel matrix (Fig. 11(c)). Furthermore, a heavily deformed Mn(S,Se) layer of approximately 50 to 60 nm in width was also found at the interface of inclusion and steel matrix (Fig. 11(d)). The interfacial crack and heavily deformed layer observed by TEM maybe indicate the loss of a coherent bond between the complex Mn(S,Se) inclusions and the steel matrix during machining.
TEM images of FIB cut sample Mn(S,Se) inclusion (a), diffraction pattern from inclusion (b), and inclusion/matrix boundary (c, d).
The rake surfaces of the worn tools were characterized using an SEM/EDS to investigate the possible deposits on the tools. Figure 12 shows the morphologies of the worn tools used for machining the Se-treated #2A and the base #2B QT steels. In both of the machining tools, two types of deposits were found. The tool used for machining the Se-treated steel #2A had a complex Mn(S,Se) deposit, and the tool used for machining the base steel #2B had a pure MnS deposit. Both types of deposits were located at the end of the cutting edge, which indicated that the manganese sulfoselenide and sulfide inclusions were truly involved in the machining process. The other type of deposit found on the worn tools was Mn–Al–Si–Ca oxide. These complex oxides were located at the outside of the cutting area.
Morphology of the worn tools used for machining steel #2A (a, b) and steel #2B (c, d). (Online version in color.)
The experimental results observed in this study revealed that the effect of Se micro-alloying on improving steel machinability strongly depended on metal matrix. Se micro-alloying was more efficient in the QT steel with high hardness, while there was no detectible effect on the machinability of HR steel. Multiple mechanisms could be involved in these processes which will be discussed based on the experimental observations provided in this article. Due to the extremely low solubility of Se in steel matrix, especially in Cr, Ni, Mo alloyed 4140 steel, the influence of Se on steel machinability is mainly dependent on the characteristics of non-metallic inclusions formed in steel.16) The set of factors and their possible mutual interactions during machining of HR and QT steels are schematically shown in Fig. 13.
The set of possible interactions during machining of Se-treated 4140 steels with different types of metal matrix. (Online version in color.)
With respect to adding Se to steel, one of the most important effects is to change the composition and morphology of sulfide inclusions. It has been reported that Se promotes the spheroidization of MnS inclusions.18) This effect was observed in the present investigation, as presented in Fig. 4. The aspect ratio of sulfide inclusions was reduced with increasing the Se/S ratio (Fig. 10). The shape of sulfide inclusions is reported to have a considerable effect on steel machinability.22) Elongated strip-like sulfides are easier to deform under the pressure of the surrounding metallic matrix, while spherical sulfides are less likely to be deformed. Owing to the non-uniform deformation of globular sulfides, non-uniform stress distribution will form along the interface between inclusions and steel matrix, thus providing favorable conditions for the nucleation of cracks.23) Solid solution strengthening of sulfide inclusions by Se was inferred in this TEM study (Fig. 11). Changing sulfide inclusion shape and strengthening the sulfide would both be beneficial to chip breaking during machining. This effect could be more prominent in steels with a hard matrix, such as in the QT condition.
The second group of factors that could affect tool wear is related to tool tribology. Two possible principal influences of Se-addition need to be considered: (i) decreasing tool wear by the lubrication of cutting surface with the inclusion deposit and (ii) the negative abrasive effect caused by hard abrasive inclusions, which will increase tool wear. The observation of the worn tool used for machining Se-treated QT steel showed that the lubrication layer formed on the tool surface contained complex Mn(S,Se) (Fig. 12). The formation of lubrication layer is a complex process and depends on non-metallic inclusion composition, particle shape, metal matrix hardness as well as temperature on cutting surface. The temperature in cutting zone is higher when QT steel with a high hardness is machined. Higher temperature will soften Mn(S,Se) inclusion and promote forming a lubrication layer of between the cutting tool and the surface being machined and thus reduce tool wear.19) However, in HR steel, Se strengthening of non-metallic inclusions could increase abrasive tool wear and meditate the positive effects of the intensification of chips breaking and tool lubrication.
The effect of Se on the machinability of AISI 4140 steel with different conditions of metal matrix (in as-hot rolled (HR) and quenched & tempered (QT) conditions) was investigated. The following conclusions were drawn:
(1) Regarding the HR condition, Se micro-alloying did not show a measurable improvement in steel machinability. On the contrary, the base HR steel exhibited slightly lower tool flank wear which could be explained by the abrasive effect of the harder complex Mn(S,Se) inclusions formed in the Se-treated HR steel.
(2) The Se-treated QT steel exhibited superior machinability compared to the base QT steel.
(3) The difference in machinability of AISI 4140 steel with different processing indicates that the effect of Se on improving steel machinability is strongly dependent on metal matrix and that Se is only effective in 4140 steel with high hardness. One possible explanation for this observation is that higher temperatures develop in the cutting zone when steel with high hardness is machined, promoting a lubricant film of Mn(S,Se) at the interface between cutting tool and shavings and thus reduce tool wear.
(4) The influencing mechanism of Se on steel machinability is proposed to be the mutual effects of the globulariztion of MnS inclusions, the absence of a coherent bond between Mn(S,Se) inclusions and steel matrix, and also the formation of lubricant film in the contact zone of tool and the machining chip.
The authors would like to thank the industrial sponsors of the Kent D. Peaslee Steel Manufacturing Research Center for funding this research. The authors also acknowledged the China Scholarship Council (201806460049).
