Workpiece engineering to improve surface quality S.V. Subramanian , H.O.Gekonde and X.Zhang Department of Material Science and Engineering McMaster university, Hamilton, Ontario, Canada

ABSTRACT.

Quantitative understanding of the micro-mechanisms of tool-workpiece interaction is essential in order to improve tool life and surface quality in high speed finish machining. Investigations on the tool wear mechanisms have confirmed that chemical wear dominates at high cutting speeds due to high temperature caused by shear localisation in the primary and secondary shear zones. Microstructural parameters, i.e., matrix hardening and second phase particles, as well as metal cutting variables, i.e., cutting speed and feed, influence shear localisation in the primary shear zone. The temperature rise at the cutting edge of the tool due to shear localisation in the primary shear zone causes chemical wear by oxidation of the cutting edge of the tool. By engineering reactive elements with strong affinity for oxygen in the workpiece, it is shown that the cutting edge of the tool is protected by forming in-situ a stable refractory oxide layer, which suppresses chemical wear. Shear localisation in the secondary shear zone is due to seizure occurring at the tool-chip interface. This raises the local temperature at the tool-chip interface, causing chemical crater wear by diffusion mechanism. By engineering glassy inclusions into the workpiece that form a viscous layer of adequate thickness to lubricate tool-chip interface, seizure is prevented and consequently chemical crater wear is suppressed. These models are validated with experimental results, which demonstrate that workpiece engineering is an integral aspect of surface quality control in high speed machining. KEY WORDS:

Tool chemical wear, high speed machining, surface quality.

1. Introduction High speed machining has generated great technological interest because it has the potential to offer excellent surface finish under dry machining conditions, increase productivity and decrease cost. In finish turning operations, high speed coupled with low feed rates have been used successfully to achieve excellent surface finish so that a subsequent finish grinding operation is eliminated, resulting in substantial cost

savings. However, an outstanding problem is the chemical wear of the tool occurring at the cutting edge caused by workpiece variables. Extensive research has been carried out over the years to get a quantitative understanding of mechanisms underlying chemical wear and its relationship to workpiece variables [TRE 84, SUH 86, KRA 80, ING 93, SUB 93, RAM 94 and GEK 98]. Temperature and pressure at the tool-chip interface determine chemical wear due to tool-workpiece interaction at high cutting speed. These are strongly influenced by shear localisation in metal cutting. Figure1(a) is a schematic, showing shear localization occurring in the primary and secondary shear zones in a chip from large strain, high strain rate deformation in metal cutting, and Figure 1(b) the consequent tool wear caused by shear localization in the primary and secondary shear zone respectively. The consequence of high temperature due to shear localization in the primary shear zone is chemical wear initiated by oxidation of the cutting edge of the tool. By contrast, the consequence of high temperature due to shear localization in the secondary shear zone is chemical crater wear on the rake face of the tool, located at some distance away from the cutting edge of the tool. Shear localisation in the primary shear zone invariably promotes change in chip morphology from flow type to partially or fully segmented chip morphology [GEK 98]. Both microstructural parameters and metal cutting variables govern shear localisation in the primary shear zone. The flow stress behavior of work piece in metal cutting determines shear instability and hence the tendency for shear localisation in the primary shear zone. The flow stress behavior of the metal is influenced by work-hardening capacity of the alloy, the matrix hardening (thermal softening potential), second phase particles (geometric softening potential), and temperature corrected strain rate (cutting speed and feed) [SUB 99-1]. The flow stress data of the work piece material at large strain, high strain rate characteristic of machining are difficult to obtain, but this data-base is essential for predicting chip morphology [TUR 82]. The flow chip morphology exhibits a constant shear angle. The variation in chip thickness occurring in serrated chip morphology is associated with variation in shear angle within each periodic cycle and this is attributed to shear instability [KOM 82]. Shear angle in the primary shear zone is a free boundary problem and is influenced by the very factors that cause shear instability[SUB 99-1]. It is also influenced by the tribological conditions at the tool-chip interface, though this is a second order effect. Shear localisation in the secondary shear zone is caused by seizure at the tool-chip interface [TRE 88]. The origin of shear localisation and its consequences on chip formation, tool wear and surface finish in metal cutting, were investigated in a range of model iron alloys [GEK 98]. Based on the micro-mechanisms of tool wear, the concept of workpiece engineering is developed in order to suppress chemical wear during high speed machining. By engineering reactive elements with strong affinity for oxygen in the workpiece, the cutting edge of the tool is protected by forming in- situ a stable

Tool rake face

Idealised secondary shear zone

Idealised primary shear zone

Chip

Tool Tool flank face

Feed Workpiece

(a) Nose radius Chemical wear due to high temperature in the primary shear zone

Rake face of the tool Crater wear due to high temperature in the secondary shear zone

Wear land on flank face due to physical wear Cutting edge Flank face of the tool (b) Figure 1. (a) Schematic representation of primary and secondary shear zone, occurring in the chip during the metal cutting and (b) Schematic representation of tool wear caused by shear localisation in the chip during metal cutting. Temperature rise caused by shear localisation in the primary shear zone localises the chemical wear at the cutting edge of the tool. Temperature rise in the secondary shear zone caused by seizure at the tool-chip interface localises the crater wear at some distance from the cutting edge of the tool.

refractory oxide layer, which suppresses chemical wear. The preservation of the cutting edge of the tool ensures consistent surface quality. By engineering glassy inclusions into the workpiece that form a viscous layer of adequate thickness at the tool-chip interface which lubricate tool-chip interface, it is demonstrated that chemical crater wear is suppressed and hence the tool life is prolonged [SUB 97 and SUB 98-1].

2. Analysis of Shear Localisation Theoretical analysis and experimental investigations on shear localisation in metal cutting were carried out in a range of iron alloys with varying matrix hardening and volume fraction of inclusions [GEK 97-1 and SUB 98-2]. The microstructural changes in the chips were characterised using optical metallography, SEM, TEM and X-ray diffraction techniques. Contact length and crater wear were characterized using optical and scanning electron microscopy. The force measurements were carried out. Chemical wear was determined using Instrumental Neutron Activation Analysis (INNA), Inductively Coupled Plasma Mass Spectrometry (ICPMS) and Secondary Ion Mass Spectrometry (SIMS). Salient observations relating to shear localisation and its consequence on chip morphology, tool wear and surface finish are presented and discussed.

2.1. Shear Localisation in Primary Shear Zone in Metal Cutting Matrix and second phase particles are important microstructural parameters that influence shear localisation. Thermal softening potential of the matrix, i.e., the change in flow stress with temperature causes thermoplastic shear localisation in accordance with the mechanism originally proposed by Zener and Hollomon about fifty years ago [ZEN 44]. When an increment in stress due to strain hardening of the matrix is overcome by a decrease in stress due to thermal softening, plastic deformation becomes unstable and the homogeneous deformation gives way to a localised band-like deformation to form adiabatic shear bands. The incompatibility of deformation between the second phase particles and the matrix causes void nucleation and growth and heterogeneous deformation leading to ductile fracture. The flow localisation caused by the instability strain in the presence of voids is referred to as geometrical softening. The general prediction for the onset of instability is defined in the literature by either in terms of a maximum in shear stress or the corresponding strain, referred to as critical strain for instability. Bai and Dodd [BAI 92] have concluded that instability is a necessary but not an adequate condition for localisation. Meyers [MEY 95] has postulated that shear localisation is caused by a sharp decrease in flow stress accompanying a major microstructural event such as dynamic recrystallisation. Dodd and Atkins [DOD 83] have analysed flow localisation in shear deformation under condition of thermal softening, geometrical softening and a combination of both. This approach has been extended in the present work on iron alloys to analyse shear instability and shear localisation occurring in metal cutting [SUB 98-2]. The change in flow stress accompanying phase transformation is shown in the present work on Fe-Ni-C model alloys to cause shear localisation in metal cutting [GEK 98]. 2.1.1. Effect of Matrix Hardening on Shear Localisation and Its Consequences on Tool Wear

Studies on the influence of thermal softening potential of the matrix on shear localisation was carried out in Fe-28%Ni-0.1%C alloy, heat treated to obtain a fully martensitic structure. Shear localisation is revealed metallographically by the white transformation band formed by the reverse transformation of martensite to austenite. Figure 2(a) shows continuous chip morphology obtained at a cutting speed of 25 m/min, with the shear localisation occurring in the secondary shear zone at the toolchip interface as delineated by the white band, and Figure 2(b) shows the corresponding tool crater wear which can be barely detected. On increasing the cutting speed to 75 m/min but keeping the feed constant at 2.5 mm/rev, the chips exhibit the onset of shear localisation in the primary shear band in addition to transformation shear band in the secondary shear zone. Significant crater wear occurs but the crater is located at some distance from the cutting edge. On increasing the cutting speed to 150 m/min, the chip exhibits shear localisation both in the primary and secondary shear zone as shown in Figure 3(a), and the crater draws closer to the cutting edge, see Figure 3(b). The white band in the picture is the region that has undergone phase transformation from martensite to austenite. On annealing the alloy to fully austenitic condition before machining, the chip reverts to fully continuous chip morphology, which demonstrates that in the absence of thermal softening potential of matrix, shear localisation does not occur and hence the chip morphology tends to be flow type. Figure 4(a) shows the flow chip morphology of Fe-28%Ni-0.1%C alloy in the annealed condition at a cutting speed of 350 m/min, and Figure 4(b) shows the corresponding tool crater wear located well away from the cutting edge. Theoretical analysis shows that the temperature in the primary shear zone will increase linearly with cutting speed once the shear is localised [BAI 92]. Figure 5(a) is an optical micrograph of a fully segmented chip obtained at a cutting speed of 456 m/min, showing that transformation shear (white) band envelops the segmented chip in Fe-28%Ni-0.1%C alloy. TEM characterisation of the transformation shear band caused by shear localisation in the primary shear zone showed ultra-fine grains (less than one tenth of a micrometer), suggesting that conditions for dynamic recrystallisation are attained in the primary transformation shear band [GEK 97-1 and GEK 98]. Figure 5(b) shows severe wear at cutting edge of the tool, caused by chemical wear due to interaction of the primary shear zone at high temperature with the cutting edge of the tool. The loss of cutting edge of the tool impairs the surface finish. 2.1.2. Effect of Second Phase Particles on Shear Localisation and its Consequence on Tool Wear The effect of second phase particles on shear localisation was investigated by comparing the metal cutting behavior of ferritic ductile iron with AISI 1020 steel. Ductile iron is a model alloy for geometric softening due to a large volume percentage (~10%) of graphite nodules in a ferritic matrix whereas AISI 1020 steel in the normalised condition is a model alloy, devoid of thermal and geometric softening potential. Figure 6(a) is an optical micrograph of ductile iron chip obtained

(a) (b) Figure 2. (a) Flow chip morphology of hardened Fe-28.9Ni-0.1C alloy at cutting speed of 25m/min, exhibiting shear localisation and (b) Tool crater wear after 10 s cutting at 25 m/min. (note crater wear well removed from cutting edge)

(a) (b) Figure 3. (a) Partially segmented chip morphology of hardened Fe-28.9Ni-0.1C alloy, at cutting speed of 150 m/min, exhibiting shear localisation in the primary and secondary shear zone and (b) Tool crater wear after 10 s cutting at 150 m/min.

(a) (b) Figure 4. (a) Flow chip morphology of annealed Fe-28.9Ni-0.1C alloy at cutting speed of 350 m/min and (b) Tool crater wear after 10 s cutting at 350 m/min.

at a cutting speed of 300 m/min. Chip segmentation is caused by geometrical softening due to a large volume fraction of graphite nodules in ferritic matrix. Figure 6(b) shows the localisation of the crater at the cutting edge of the tool caused by shear localisation in the primary shear zone. The crater wear on the tool is localised at the cutting edge. In contrast to ductile iron, AISI 1020 steel exhibited flow chip morphology under the same cutting conditions. Even though shear localisation was absent in the primary shear zone, the chips did exhibit shear localisation at the toolchip interface caused by the tribology of seizure at the tool chip interface temperature. In consequence, crater is formed on the rake face of the tool but the crater is located well away from the cutting edge of the tool. Figure 7 shows the tool crater wear obtained on machining in AISI 1020 steel and the crater is located away from the cutting edge of the tool.

(a) (b) Figure 5. (a) A nearly fully segmented chip of hardened Fe-28.9Ni-0.1C alloy at a cutting speed of 456 m/min and (b) Interaction of the primary shear zone at high temperature with cutting edge, resulting in loss of cutting edge of tool at 456 m/min.

(a) (b) Figure 6. (a) A fully segmented chip of ferritic ductile iron at cutting speed 300 m/min (shear localisation in the primary shear zone caused by geometrical softening) and (b) Tool crater wear, crater localised at cutting edge of tool caused by shear localisation in the primary shear zone.

Figure 7. Crater wear observed on the cemented carbide tool (K 1) after 20 seconds on machining AISI 1020 steel at cutting speed of 300 m/min.(note the crater occurring farther away from the cutting edge of the tool.

Once shear localization occurs in the primary shear zone, the work of deformation is localized in a narrow region, when the bulk (85-95%) of the work due to plastic deformation of metal converts into heat. At high strain rates characteristic of high cutting speeds, there is inadequate time for the heat to escape from the shear zone, which results in local temperature rise in the primary shear zone. An important consequence of shear localization at high strain rate deformation is steep temperature rise in the primary shear band irrespective of whether shear localization is caused by thermal softening due to a hardened matrix or geometric softening due to second phase particles. The interaction of the transformation shear band at high temperature with the cutting edge of the tool raises the temperature at the cutting edge of the tool. This, in turn, causes chemical wear at the cutting edge of the tool. 2.1.3. Effect of metal cutting variables on shear localisation in primary shear zone The effect of cutting speed, feed, depth of cut and coolants were studied on Fe28%Ni-0.1%C alloy. The microstructural changes accompanying phase transformation were used to map shear localisation. Cutting speed and feed are found to be important metal cutting variables that influence shear localisation. The effect of increasing the feed is to promote shear localisation in the primary shear zone, leading to chip segmentation. In finish machining, a low feed is used. On reducing the feed from 0.397 to 0.055 mm/rev, the shear localisation in the primary shear zone is suppressed, and the chip morphology changes from fully segmented to flow type at a cutting speed of 350 m/min in hardened Fe-28%Ni-0.1%C alloy, see Figures 8(a) and (b). Even though shear localisation in the primary shear zone is suppressed by decrease in feed, shear localisation in the secondary shear zone

persists, which results in chemical crater wear. With the decrease in feed, the contact length decreases, drawing the crater close to the cutting edge of the tool.

(a) (b) Figure 8. (a) Full segmentation of chip caused by large feed (0.397 mm/rev) at cutting speed of 350 m/min and (b) Flow chip morphology caused by low feed (0.055 mm/rev) at the same cutting speed of 350 m/min.

3. Consequence of shear localisation in the primary shear zone 3.1. Chemical Wear at The Cutting Edge of The Tool 3.1.1. Studies on Mechanisms of Chemical Wear at The Cutting Edge of The Tool Extensive studies were carried out to quantify the mechanism of tool wear at the cutting edge of the tool during finish machining of pearlitic gray cast iron using cubic boron nitride tool over a range of cutting speeds and feeds. The graphite flakes are A-type and are randomly oriented but interconnected in a fully pearlitic iron matrix. The large volume fraction of graphite flakes facilitates shear localisation in the primary shear zone due to geometrical softening effect. Figure 9 is a typical chip obtained at a cutting speed of 2,194 m/min at a feed of 0.15mm/rev using cubic boron nitride tool. The chip is fully segmented due to geometric softening due to large volume fraction of graphite flakes and in consequence chemical wear is expected at the cutting edge of the tool. Figure 10 is an SEM picture of the cubic boron nitride tool showing extensive tool wear localised at the cutting edge of the tool after finish machining of as few as five pearlitic gray iron castings. Since the wear extends well into the flank face of the tool, the origin of the wear was mistaken in the past for physical wear processes. The surface finish deteriorated after machining very few (five) castings. By engineering reactive elements with strong affinity for oxygen in the matrix of the workpiece, the cutting edge of the tool is

Figure 9. An optical micrograph of a polished section of a typical chip obtained in high speed finish machining of gray cast iron at a high cutting speed of 2,194 m/min and a feed of 0.15mm/rev. The chip mounted in bakelite shows fully segmented chip morphology. The dark long A-type graphite flakes are distributed randomly in bright iron matrix. The segmentation of the chip is due to localisation of shear in the primary shear zone. The consequent temperature rise accelerates chemical wear at the cutting edge of the tool. protected against chemical wear. Figure 11 is a SEM picture of the cubic boron nitride tool after machining 300 pieces of casting at the same cutting speed of 2,194 m/min and feed of 0.15mm/rev. The cutting edge of the tool was well preserved even after machining 300 pieces and hence the surface finish obtained was consistently good. Energy dispersive analysis and SIMS analysis of the cutting edge of the protected layer formed by in-situ reaction with reactive elements from work piece engineered casting showed evidence for the presence of B and N in addition toSi, Al and O. In comparison, the tool which exhibited severe chemical wear at the cutting edge showed conspicuous absence of B and N of the tool material but the presence of Fe, Si and O, indicative of unstable, low melting oxides.

3.2. Method of prevention of chemical wear at the cutting edge of the tool According to the proposed mechanism, the temperature of the cutting edge of the tool is raised by shear localisation in the primary shear zone. In the presence of air, the cutting edge is oxidised. Diffusional wear is accelerated by high diffusivity of oxygen in the liquid phase formed by low melting oxides. By engineering a reactive element with strong affinity for oxygen in the cast iron matrix, the unstable oxide layer is reduced by the reactive element to form a stable protective refractory oxide layer by in-situ reaction at the tool-chip interface. The diffusional wear is retarded by several orders of magnitude by the formation of the chemically stable high

melting (refractory) oxide layer on the tool surface capable of protecting the tool material from further oxidation. Thermo-kinetic models have been developed to engineer a small amount of reactive elements with high affinity for oxygen dissolved in iron matrix, designed to reduce the unstable oxides and form a stable refractory oxide layer. The stable oxide layer protects the tool from further oxidation, while suppressing chemical wear by diffusion mechanism by several orders of magnitude.

Figure 10. A SEM picture of typical wear obtained during high speed machining of gray cast iron with cubic boron nitride tool at 2,194 m/min at a feed of 0.15 mm/rev. The wear localised at the cutting edge of the tool is chemical in origin and is caused by high temperature arising from shear localisation in the primary shear zone.

Figure 11. A SEM picture of the cubic boron nitride tool after machining 300 castings at a cutting speed of 2,194 m/min at a feed of 0.15mm/rev. The cutting edge of the tool is well preserved by a protective layer formed on the tool surface by tool– workpiece interaction.

3.3. Workpiece engineering to prevent chemical wear at the cutting edge of the tool The concept of workpiece engineering is to design the chemistry of the workpiece material in order to preserve the cutting edge of the tool from chemical wear. Reactive elements with strong affinity for oxygen are engineered in the iron matrix, which are designed to reduce any unstable oxides formed at the cutting edge of the tool, and to form in-situ a chemically stable refractory oxide during metal cutting. The protective layer is designed to act as a diffusion barrier at high temperature caused by shear localisation in metal cutting. The concentration of reactive elements required in the iron matrix to suppress chemical wear is exceedingly small, typically less than 100 ppm (