Keywords

1 Introduction

Nickel-based superalloys have an exceptional combination of high-temperature strength, toughness, and resistance to degradation in corrosive and oxidizing environments. Therefore, these alloys become an ideal material choice for use in the hot sections of gas turbine engines for aerospace and power generation applications [1]. A noteworthy feature of nickel-base alloys is that they can be used for load-bearing applications at temperatures up to 80% of melting temperature (1000 °C) of the alloy [2].

The major phases present in nickel-based superalloys are γ and γ'. The continuous matrix, γ is a face-centered cubic (FCC) nickel-based austenitic phase that contains a high percentage of high strength and refractory elements. This matrix has high mechanical properties by the solid solution strengthening and high modulus. The primary strengthening phase, γ' is Ni3(Al/Ti) coherently precipitating phase with an ordered L12 (FCC) crystal structure. Interestingly, the flow stress of the γ' increases with increasing temperature up to 650 °C due to Kear Wilsdorf lock [3]. In addition, γ' is quite ductile and thus imparts strength to the matrix without lowering the fracture toughness.

Nickel-based superalloys can be cast into components of complex shapes and/or internal configurations with metallurgically controlled uniform microstructure. Vacuum investment casting process is employed to produce cost-effective components with a high degree of dimensional accuracy, surface finish, and structural integrity [4]. Initially, the gas turbine blades were made as polycrystal (equi-axed) castings; later, development of the columnar-grained castings resulted in some significant improvements in creep strength and thermal fatigue resistance. Advanced cast, single-crystal (SC) nickel-based superalloy castings offer further improvement of high-temperature capabilities and resulting in superior turbine engine performance and durability [5]. The SC castings thus produced by the vacuum investment casting process undergo various machining operations to obtain fitment geometries for the engine assemblies. The nickel-based superalloys are considered as hard and difficult-to-cut materials due to high strength, hot hardness, fracture toughness and work hardening, and low thermal conductivity [6]. Higher cutting forces and heat are generated while machining these alloys, which result in premature failure of cutting tools and poor integrity of the machined surfaces [7].

The surface integrity (SI) of a machined component is the sum of all the elements that describe the conditions existing on or at the surface. The term was coined initially by Field and Kahles in the year 1964 [8]. The formation of a white layer at the machined surface is attributed to severe plastic deformation, grain refinement, phase transformation under rapid heating and quenching [9,10,11,12]. The SI aspects of the component are of prime concern in aerospace applications as the failures are primarily caused due to fatigue, creep, and stress corrosion, etc. [13].

Although there have been numerous studies on different aspects of cutting process over the years on superalloys, the properly published literature on SI aspects in the machining of SC nickel-based superalloys is not available. The SC superalloys are highly sensitive to thermo-mechanical loads. Hence, a systematic experimental study has been carried on a commercial second-generation single-crystal nickel-based superalloy, CMSX-4 superalloy by using various precision conventional and unconventional machining process, such as vertical milling, surface grinding, wire electric discharge machining (WEDM), and abrasives water jet machining (AWJM), respectively, which induces certain thermo-mechanical loads.

This paper focuses on the surface topography and surface metallurgy of the machined surfaces. The overall theme of the experimental study with input parameters and output responses is presented in Fig. 1.

Fig. 1
A workflow of the experimental investigation in three steps. It includes input parameters, process, and output response.

Overall theme of the experimental investigation

Full size image

2 Experimental Setup

2.1 Work Material

The single-crystals of the superalloy were supplied by Cannon Muskegon Corporation, USA, in the form of round rods with axis parallel to [001] direction. The nominal composition of CMSX-4 superalloy by weight percent is shown in Table 1.

Table 1 Nominal chemical composition of CMSX-4 alloy in wt.%
Full size table

This alloy maintains its ultra-high strength even at high temperatures by solid solution strengthening and precipitation hardening through the balanced alloy chemistry [14]. Rhenium element in this alloy mainly partitions into the γ matrix, which retards coarsening of γ' strengthening phase and increases γ/γ' misfit. Therefore, this alloy exhibits good resistance to excessive γ' growth during service. The physical and mechanical properties of CMSX-4 alloy from the supplier technical data are shown in Table 2.

Table 2 Physical and mechanical properties of CMSX-4 alloy
Full size table

2.2 Specimen Preparation

Solid casting blocks were cut by a precision WEDM process from the fir tree side of SC turbine blade castings, which were manufactured with <001> grain growth direction from the prefix bottom seeding and directional solidification technique through vacuum investment casting process. The test specimens were of length of 47 mm, width of 30 mm, and thickness of 12 mm. The micrographs of transverse to the grain growth direction of the base metal in as-cast condition are shown in Fig. 2. The optical micrographs reveal that the casting has four-fold dendrite core morphology and eutectic phases in the inter-dendrite region. These eutectic phases can be seen as bright lamellar regions in the inter-dendrite region. The high magnification SEM images (Fig. 2b) of the dendritic core show the uniform distribution of γ’ phase embedded in the γ matrix.

Fig. 2
Two micrographs of transverse to the grain growth direction of the base metal in as-cast condition under 50 times and 50,000-time magnification.

Micrographs of transverse to the grain growth direction of the CMSX-4 alloy in as-cast condition a Optical under 50× magnification and b SEM under 50,000× magnification

Full size image

2.3 Machining Experiments

It is known that the milling and grinding process is conventional machining process, wherein the cutting action carries out by the combination of thermal and mechanical loads. The WEDM process is an unconventional machining process, wherein the cutting action carries out by the thermal erosion. The AWJM process is also an unconventional machining process, wherein the cutting action is carried out by the mechanical erosion from the high kinetic energy of water and abrasives. The thermo-mechanical loads in qualitative and specific energy in quantitative manner of the subjected machining process are shown in Fig. 3.

Fig. 3
Two graphs depict the thermomechanical loads in qualitative and specific energy in a quantitative manner of the subjected machining process.

Precision machining processes a thermo-mechanical loads and b semi-log plot of specific energies

Full size image

Milling Process: A knee-type conventional vertical milling machine (FN2V model from M/s HMT, India) of spindle motor power of 5.5 kW and rotational speed of 1800 per min was used. The sample was firmly fixed on the table, and face milling operations were performed with the solid carbide insert number TPN 2204 PDR 3040 (M/s Sandvik Coromant, Pune) fitted into a 100-mm-diameter cutter of 8 flutes. The flood coolant with a stable emulsion of 1:20 water ratio was employed with cutting speed of 50 m/min, table feed of 16 mm/min, and depth of cut of 0.2 mm.

Grinding Process: Grinding experiments were performed by a conventional horizontal spindle-type surface grinding machine (PRAGA 452P model from M/s HMT, India). The power rating of the spindle drive was 1.5 kW, and the maximum speed was 2800 rpm. A conventional tool room-type (white color) alumina-abrasive wheel of specification A60K5V8 (M/s Carborundum Universal Limited, Chennai) was used. The wheel has an outer diameter of 200 mm, inner diameter of 31.75 mm, and width of 20 mm. The table speed of 1.7 m/min, wheel speed of 1540 m/min, and depth of cut of 0.03 mm were used along with the soluble cutting oil, Servocut S grade with 3% concentration.

WEDM Process: The cutting experiments were conducted with 0.25-mm-diameter hard brass wire by a submerged-type precision WEDM machine (CNC Ultracut S2 model from M/s Electronica Machine Tools, Pune). Gap voltage of 36 V, current of 1.4 A, time ON of 10 µs, cutting speed of 0.5 mm/min, and wire tension of 11 N were used along with the de-ionized water with total dissolved salts of below 3 ppm as dielectric.

AWJM Process: The cutting experiments were conducted with 80 mesh size garnet abrasives and 0.7 mm diameter sappier orifice nozzle by a water jet machining center (55,100 model from M/s Omax, USA). The parameters of surface quality of 5 number (i.e., high), water jet pressure of 255 MPa, abrasive flow rate of 340 g/min, traverse speed of 35 mm/min, and stand-off distance of 3 mm were used during the experiments.

2.4 Characterization Techniques

The surface roughness measurements were carried out on the machined surface by using a contact-type instrument (Form Talysurf Intra model from M/s Taylor Hobson, UK) with a 2 µm conisphere-shaped diamond stylus at 3 different locations along the machining direction. A cross-section of 12 mm × 12 mm with 10 mm thick was extracted from the machined surfaces of each machining process by using a precision WEDM. These geometries were mounted into the 1-inch size Bakelite molds and polished using a series of silicon carbide emery papers and diamond paste. The cleaned surfaces were etched by a commercial Kalling’s 2 regent. The micro-structural analysis was performed by an optical microscope (DM 4000 M model from M/s Leica Microsystems GmbH, Germany) and a scanning electron microscope (EVO-18 SEM model from M/s Carl Zeiss, Germany). The white-layer thickness was measured on the digital micrographs by using an image processing software (Image J). The recrystallization, crystal orientation, and mis-orientation angle studies were carried by using the electron backscatter diffraction (EBSD) technique (Gemini SEM 300 model from Carl Zeiss GmbH, Germany). The hardness on the machined surfaces was measured by using a micro-Vickers hardness instrument (MMT XZ model from M/s Matsuzawa, Japan) with diamond pyramid indenter under 0.5 kg load and dwell time of 10 s.

3 Results and Discussions

Optical images of the cross-section of the machined surfaces in Fig. 4 show that the machined surfaces have not consisted of micro-cracks.

Fig. 4
Four optical images exhibit the cross-section of the machined surfaces, milled surface, ground surface, W E D M surface, and A W J M process.

Optical images (200X) of the machined surfaces of a milled, b ground, c WEDM and d AWJM process

Full size image

The surface machined by WEDM has highest surface irregularities, and AWJM has the highest waviness.

3.1 Surface Roughness

The results of the average surface roughness are reported in Fig. 5. It is observed that the ground surface has the highest surface finish due to the cutting and polishing action of abrasives. The WEDM surface has the lowest surface finish due to higher surface irregularities by the discrete spark erosions and deposition of debris.

Fig. 5
A bar graph depicts the average surface roughness in micrometers. The surface roughness is 0.7 for milled, 0.2 for ground, 3.2 for W E D M, and 2.4 for A W J M.

Average surface roughness of the machined surfaces under subjected conditions

Full size image

The milled surface has a better surface finish due to the engagement of multiple cutting inserts.

3.2 White-Layer Thickness

The SEM images of the cross-section of the machined surfaces are shown in Fig. 6.

Fig. 6
Four S E M images and a bar graph of milled, ground, W E D M, and A W J M surfaces after the machining process exhibit that A W J M has the highest white-layer thickness.

SEM images (10,000×) of machined surfaces of a milled, b ground, c WEDM, d AWJM, and e results of white-layer thickness

Full size image

It can be seen that the machined surface by AWJM has the highest white-layer thickness due to higher mechanical loads while cutting, whereas the milled surface has the lowest value due to lower specific energy.

3.3 Recrystallization Depth

The results of the recrystallization studies on the machined surfaces by EBSD technique are shown in Fig. 7.

Fig. 7
Four plot depicts the results of the recrystallization studies on the milled, ground, W E D M, and A W J M surfaces. A significant deform can be observed in the milling process.

Recrystallization results of a milled, b ground, c WEDM and d AWJM surfaces

Full size image

It is observed that none of the machining process could generate recrystallization defects under applied working conditions; however, a significant deformation (shown in red color at the location of 3rd column in the figure) was observed for the milling process due to multidirectional cutting forces.

3.4 Crystal Orientation Direction

The machined surfaces were further analyzed by EBSD technique for crystal orientation mapping. The results along with inverse pole figure (IPF) color key are shown in Fig. 8.

Fig. 8
A crystal orientation mapping of the machined surfaces. The inverse pole figure color key exhibits that the machined surface has a crystal orientation equal to 001.

Crystal orientation results of a milled, b ground, c WEDM and d AWJM surfaces of CMSX-4 superalloy, and e IPF key

Full size image

It was observed that the machined surface has the crystal orientation equal to <001> direction; it was the same as that of the base metal casting.

3.5 Misorientation Angle

The EBSD results of the distribution of grain boundary misorientation angles are shown in Fig. 9. It was observed that the surfaces under all machining process have a higher fraction of misorientation angle in the range from 0 to 5° only. AWJM surface has negligible distribution beyond 0 to 5° range, whereas the WEDM surface has a small number of misorientation angles beyond 5° and up to 15°.

Fig. 9
A bar graph depicts that the surfaces under all machining processes have a higher fraction of misorientation angle in the range from 0 to 5 degrees.

Misorientation angles of the grain boundaries of milled, ground, WEDM and AWJM surfaces of CMSX-4 superalloy

Full size image

The spread of WEDM surface could be due to the high temperature exposures by the discrete electric sparks. Formation of the crack-free surface was attributed to low misorientation angles.

3.6 Machining Affected Zone

The deformation images of the machined surface by EBSD technique are shown in Fig. 10. It can be seen that the milled surface has the highest deformation zone up to the maximum depth of 100 µm from the surface, whereas the WEDM has the lowest deformation depth of 10 µm.

Fig. 10
Four figure depicts the surface deformation after milling, grinding, W E D M and A W J M process. A bar graph exhibits that the milled surface has the highest deformation depth.

Machining affected zone of CMSX-4 superalloy under a milling, b grinding, c WEDM and d AWJM process

Full size image

The ground surface has the lower deformation depth than the milled surface. The ground surface images reveal the existence of slip lines below the machined surface. This could be due to localized thermo-mechanical loads.

3.7 Surface Hardness

The average surface hardness values of the machined surfaces are shown in Fig. 11a.

Fig. 11
Two bar graphs depict the degree of surface hardness of the machined surfaces of milled, ground, W E D M, A W J M in terms of two scales Vickers hardness and degree of surface hardening.

Surface hardness of the machined surfaces of CMSX-4 alloy under various machining process a average Vickers hardness and b DSH

Full size image

The degree of surface hardening (DSH) in percentage was calculated by using the below equation:

$$\mathrm{DSH}= \frac{(\mathrm{MSH}-\mathrm{BSH})}{\mathrm{BSH}} \times 100$$
(1)

where MSH is the surface hardness of the machined surface and BSH is the surface hardness of the base metal in VHN. The results of DSH are shown in Fig. 11b. It can be seen that the machined surfaces have less than 10% DSH.

The WEDM has a higher DSH due to highest thermal loads during the cutting. The AWJM has a lower DSH though it has highest specific energy. This could be due to the direction of dominant cutting force in parallel to the machined surface.

4 Conclusions

Machining experiments were conducted on a 2nd-generation single-crystal nickel-based superalloy, CMSX-4 alloy in as-cast condition by milling, grinding, WEDM and AWJM process, which induces certain thermo-mechanical loads of various order. The following conclusions are drawn from the experimental study.

  • The ground surface has achieved the best surface roughness of 0.2 µm Ra due to the cutting and polishing actions by the abrasives, whereas the WEDM surface has achieved the poor surface roughness of 3.2 µm Ra due to the deposition of discrete micro-debris on the machined surface. The machined surface of AWJM has highest waviness due to the divergence of the water jet.

  • The specific energy is attributed to the formation of white-layer thickness. The AWJM surface has the highest value due to highest specific energy (mechanical load), and the milled surface has the lowest due to least specific energy.

  • Though the milled process has lower specific energy, the milled surface has the highest machining affected zone with high plastic deformation depth due to the multidirectional action of cutting forces during cutting. The AWJM surface has the least amount of machining affected zone due to alignment of dominant cutting force in parallel direction to the machined surface.

  • The thermal loads are attributed to the formation of misorientation angles. WEDM surface has a wide range of misorientation angles up to 15° by the high thermal loads, whereas AWJM has the least values of the angles.

  • The thermal loads are also attributed to the hardening of the machined surfaces The WEDM surface has the highest amount of DSH due to the formation of a recast layer under rapid heating and cooling.

  • Surface cracks, changes in the crystal orientation, and recrystallization were not observed on any of the machined surfaces under the machining conditions.

The chosen machining parameters, therefore, appear appropriate for machining the CMSX-4 superalloy, which obtained a higher integrity on the machined surfaces.