Some investigations in grindability improvement of Inconel 718 under ecological grinding

Abstract

This work explores the improvement in grinding characteristics of Inconel 718 (IN718) using soluble oil under minimum quantity lubrication and liquid nitrogen (LN₂) environments. The coolant flow rate in minimum quantity lubrication and LN₂ grinding has been regulated through indigenously developed setups. Grinding performance has been studied in terms of on-machine measured grinding forces and centre line average surface roughness (R_a). The obtained grinding characteristics have been compared with the outcomes under dry and wet grinding. Surface integrity of ground surface, wheel morphology, and chip formation characteristics has been studied using scanning electron microscope, energy dispersive X-ray spectroscopy, and atomic force microscopy. Analysis of variance has been carried out to capture the variability in the experimental data for tangential forces and R_a. The main effect of the factors and their first-order interactions have been considered, and second-order regression equations have been developed using response surface methodology. LN₂ grinding has been proved to be more efficient as it yielded lowest grinding forces, least oxidation, minimal ground surface damage and better surface integrity. The occurrence of almost circular chips in dry grinding indicates severe oxidation, whereas small C-type chips formed under minimum quantity lubrication and LN₂ conditions indicate effective cooling under these environments. The energy dispersive X-ray spectroscopy analysis of the ground surfaces also supports these results through the occurrence of the highest oxidation in dry grinding. From this work, it has been concluded that LN₂ and minimum quantity lubrication grinding offer a clean and effective means to improve grinding performance of IN718 compared to dry and wet grinding.

Keywords

Ecological grinding Inconel 718 minimum quantity lubrication grinding cryogenic grinding surface roughness

Introduction

The increased customer awareness, economic liberalisation, and globalisation have significantly contributed to the growing demand for quality products, leading to healthier competition among the producers globally. Right product at the right time at the right cost is becoming the punch line for most of the industries. Quality products at a reasonable price can be manufactured with the adaptation of the improved manufacturing processes. Grinding is one of the most versatile manufacturing processes widely used in making quality products, comprising around 25% of the total machining processes.¹ Nowadays, this process is being improvised to meet the customer’s requirement, initiating wide area of research aimed at process enhancement. Despite its importance, the productivity of grinding is relatively less, along with a high operating cost compared to other machining processes. Therefore, there exists a strong need for improving the productivity and reducing the associated cost of the grinding process. The important attributes of the improvement in grinding process effectiveness are shown in Figure 1.

Figure 1.

Schematic representation of the attributes for improvement in grinding process effectiveness.

The present work focuses on grinding process improvement through reducing the environmental load and promoting the usages of the environment-friendly grinding medium. Grinding is a high specific energy-intensive secondary manufacturing process, which can be attributed to the inherent multi-point cutting mechanism, and physical as well as geometrical parameters of the grits.² This culminates in high heat generation and eventually results in a high-temperature rise in the grinding zone. Moreover, in grinding with conventional wheels, the fraction of the total generated heat received by the workpiece is around 0.7.^2,3 Such high heat may cause thermal damages such as surface oxidation, surface cracks, unfavourable tensile residual stress generation, and microstructural alteration in the work surface.^4–6 Additionally, this high heat accumulation in the grinding zone can lead to rapid wheel wear. Therefore, to safeguard the ground surface from such thermal damages and minimise the wheel wear, the heat generated while grinding needs to be controlled and removed quickly. Quick heat removal can be accomplished through the application of suitable cutting fluid that must possess a high heat removal capacity. Along with proper cutting fluid selection, an appropriate method of introducing the cutting fluid into the grinding area is of prime importance. The most common way of cutting fluid delivery is wet cooling, in which plentiful supply of pressurised cutting fluid is provided into the grinding zone. This technique has been well appreciated and adopted by the manufacturing industries for many decades.⁷ With the passage of time, however, the researchers realised that wet cooling has many inherent limitations. In this technique, only a fraction of the cutting fluid injected is utilised in cooling and lubrication whereas a substantial portion remains unutilised. The high volume of the cutting fluid results in storage as well as disposal problems and more pumping power is required leading to substantial increase in manufacturing cost. In some estimates, wet cooling accounts for 15% to 20% of the total manufacturing cost of the component and is reported as one of the biggest contributors to the environmental pollution.^8–12

The cutting fluids often pose hazards to man, machine, and material by contaminating air, water, and soil. Water-mixed cutting fluids may cause staining, corrosion, and produce microorganisms, which may result in severe irritation to the operators, leading to diseases such as dermatitis and asthma. These issues necessitated effective usages of the cutting fluids or alternative ways, that is, adopting dry machining. Dry machining in its present form, however, may not be a solution since it has significant limitations in terms of high heat generation, and rapid tool wear, especially while machining high-strength alloys. However, stiffer pollution norms, consciousness towards green and ecological manufacturing, are enforcing the researchers to curtail the use of the cutting fluids.⁷ These scenarios have opened the avenue for other techniques with limited usage of cutting fluids and are known as environment-friendly sustainable techniques. The minimum quantity lubrication (MQL), cryogenic cooling, high-pressure cooling, and hybrid cooling are a few of these techniques.¹³

From economic and ecological considerations, MQL finds a unique position in machining domain. It has shown encouraging results in a higher heat generation process like grinding.^14–17 This technique has emerged as one of the clean and environment-friendly techniques and has been widely adopted by several manufacturing industries. The application of cryogen in grinding is another important environment-friendly grinding condition. Typically, cryogens are in the form of supercooled liquefied gases such as liquid nitrogen (LN₂), oxygen, and helium. However, in machining, LN₂ captured a significant dominance, especially in alloy steels, Ti and Ni alloy machining predominantly due to its inert behaviour and extraordinary heat dissipation capability.¹⁸ It is the most abundant, odourless, and non-toxic gas. Moreover, the direct discharge of liquid nitrogen into the atmosphere during machining processes is safe.

The industrial demand of the superalloys has increased many folds over last few decades. Ni-based superalloys are one of such superalloys getting broad appreciation and being used throughout the world due to their superior mechanical properties.^19–22 IN718, an Ni-based superalloy, was invented in the 1950s.²⁰ It happens to be an attractive material for many engineering applications with a production rate of almost 50% of all superalloys in the world.²⁰ IN718, an austenitic structure alloy, is one of the most commonly used Ni-based superalloy primarily due to its capability to withstand high temperature and severe operating conditions.²⁰ It has high strength, high ductility, and fatigue resistance over a wide temperature range (−150°C to 700°C) and hence highly used in making various salient components. Moreover, it exhibits high corrosion, fracture, and creep resistance which appears to be a characteristic of the austenitic alloys.^23–26 Due to these properties, it is widely used in chemical, nuclear, aerospace, and power industries exhibiting non-uniform strains.^{20,23,24,26–28} For such operating conditions, fatigue strength and creep strength become more important, which predominantly depends on the surface quality of the machined components. Out of the conventional machining processes, grinding is the process, which is used widely in making high surface quality components. IN718 has high toughness, severe strain hardening, low plasticity, and smaller thermal conductivity, which makes this material difficult-to-machine/grind (DTM/DTG) and possess machinability rating close to 15 on a scale of 100.²⁹

MQL aims to use a minimum amount of the cutting fluid to meet the requirements of the process. The acceptable range of the cutting fluid flow rates for MQL is about 50–500 mL/h. Hadad and Hadi³ used vegetable oil in MQL grinding of AA6061 alloy and S34700 stainless steel using SiC and alumina grinding wheels. They studied the variations in grinding forces and surface finish of the ground surfaces and found lesser grinding forces and better surface finish in MQL grinding. Sieniawski and Nadolny³⁰ presented a novel approach using centrifugal grinding fluid delivery in grinding performance improvement of steel CrV12 alloy. They found almost 10 times reduction in the cutting fluid in this method compared to wet cooling for the same grinding performance level. Reddy and Ghosh¹⁸ performed grinding of hardened AISI52100 steel under LN₂ and wet cooling with soluble oil using vitrified bonded alumina wheel. They critically investigated the wheel wear in long-duration grinding and found wear flat formation on alumina grits in the case of LN₂ grinding. This was attributed to the higher bond strength not allowing the dislodging of the worn-out grits due to much lower temperature while grinding under LN₂ environment. However, they observed much higher grinding ratio (G ratio) while grinding under LN₂ condition as compared to wet and dry grinding. Pavan et al.³¹ studied the grinding performance of IN718 under MQL mode using graphene nanoplatelets–based cutting fluid. The obtained experimental results favour the application of graphene nanoplatelets–based cutting fluid as it lowered the grinding forces, grinding temperature, and specific grinding energy. Tawakoli et al.³² investigated the effect of nozzle position, wheel structure, and wheel porosity in MQL grinding of 100Cr6 hardened steel using alumina grinding wheel. They studied the grinding performance through the measured value of grinding forces, surface roughness (R_a and R_z), and coefficient of friction (µ). They observed that coarse and porous grinding wheel were better for MQL grinding. Barczak and Batako³³ conducted grinding experiments with alumina wheel on EN8, M2, and EN31 steels in dry, wet, and MQL mode using Castrol ES1 oil. They studied the effectiveness of grinding by comparing the obtained specific grinding forces, R_a, and grinding temperature. They found lower specific forces, surface roughness, and temperature in MQL grinding compared to dry grinding. Retention of MQL oil film reduced the friction between the mating surfaces which eventually resulted in the low grinding zone temperature. Garcia et al.³⁴ used liquid CO₂ in minimum coolant grinding (MCG) of AISI4140 steel using alumina wheel. They estimated the grinding performance in terms of measurement of wear flat onto the grits, grinding forces, and specific grinding energy. Based on their studies, they obtained a much favourable grinding situation in case of MCG compared to dry grinding. This was attributed to improved lubrication in case of MCG grinding. Setti et al.³⁵ investigated the grindability of Ti-6Al-4V using LN₂ as grinding medium. They compared the obtained grinding forces, R_a, and wheel wear with dry and wet (soluble oil) grinding. In LN₂ grinding, better surface finish and lesser grinding forces were observed due to the presence of cooling mist along the wheel periphery that provided better slipping of the grits over the work surface resulting in ease of grinding. They also called cryogenic grinding as ecological grinding especially suited for DTM like Ti-6Al-4V. Ramesh et al.⁸ used chilled air mixed with a biodegradable oil in grinding of alloy steels. They developed a thermal energy equilibrium model that relates to the grinding energy, heat transfer, and internal energy rise. This model predicted the maximum material removal for chilled air grinding, which was in agreement with the experimental findings.

From the cited literature, it is evident that in the present scenario, MQL and cryogenic grinding are well accepted and appreciated cutting environments in the manufacturing sector. Till now, MQL and LN₂ grinding have been used in the grindability improvement in various steel alloys, Ti alloys, and few Ni alloys as well. However, virtually no work has been reported for MQL and LN₂ grinding of IN718. Therefore, it is of the highest importance to explore grindability of IN718 under these ecological grinding environments, which can be useful in appropriate process parameter selection. Hence, the present work is aimed at the grindability improvement of IN718 products through improvement in grinding process effectiveness by adopting environmental-friendly grinding mediums. Keeping this in purview, the objectives of the current work have been framed as follows:

Grindability study through observance of normal and tangential grinding forces, and R_a in surface grinding experiments under dry, wet, MQL, and LN₂ environments.

Study of the ground surface integrity through the finding of scanning electron microscope (SEM) study, energy dispersive X-ray spectroscopy (EDS) analysis, and atomic force microscopy (AFM) analysis.

Finding the appropriate grinding environment based on the measured grinding forces and surface integrity of the ground surfaces.

Analysis of variance (ANOVA) of the two most significant grinding indices – tangential forces and R_a.

Materials and methods

This section includes the details of experimental runs, grinding wheel, and work materials. It also contains the detailed methodology of conducting grinding experiments. The section covers the description of the responses and their measuring methodology. In the end, there is a brief description about the ANOVA of tangential forces and mean surface roughness.

Deciding the experimental runs

The number of experimental runs has been decided and customised using face-centred central composite design (FCC) which is based on the fundamentals of response surface methodology (RSM). In the present work, three factors with three levels have been selected, which gives 20 runs in each condition through FCC. All the experiments have been replicated three times, and subsequently, the calculated mean value of the responses (grinding forces and R_a) has been reported. Factors with their levels and complete details of the experimental runs are outlined in Tables 1 and 2, respectively.

Table 1.

Grinding factors and their level.

Factors	Factor level
	−1	0	1
Wheel speed, V_c (m/s)	10	15	20
Table speed, V_w (m/min)	3	9	15
Depth of cut, d (μm)	10	15	20

Table 2.

List of experimental runs in coded and uncoded form of variables.

	Run order	Coded variables			Uncoded variables
		Wheel speed	Table speed	Depth of cut	Wheel speed	Table speed	Depth of cut
Cube points	1	−1	−1	−1	10	3	10
	2	1	−1	−1	20	3	10
	3	−1	1	−1	10	15	10
	4	1	1	−1	20	15	10
	5	−1	−1	1	10	3	20
	6	1	−1	1	20	3	20
	7	−1	1	1	10	15	20
	8	1	1	1	20	15	20
Axial points	9	−1	0	0	10	9	15
	10	1	0	0	20	9	15
	11	0	−1	0	15	3	15
	12	0	1	0	15	15	15
	13	0	0	−1	15	9	10
	14	0	0	1	15	9	20
Centre points	15	0	0	0	15	9	15
	16	0	0	0	15	9	15
	17	0	0	0	15	9	15
	18	0	0	0	15	9	15
	19	0	0	0	15	9	15
	20	0	0	0	15	9	15

Work material, grinding wheel, grinding environments, and grinding kinematics

The commercially available age hardened IN718 plate with the dimensions of 70 mm (length) × 60 mm (height) × 10 mm (width) has been chosen as a work material. All the grinding experiments have been performed in down grinding using vitrified bonded white alumina wheel (A-60-K-5-V) of dimensions 350 mm (diameter) × 40 mm (width) × 127 mm (bore size). Surface grinding experiments have been conducted using two axes Chevalier Smart H1224II CNC surface grinder under dry, wet with soluble oil, MQL with soluble oil, and LN₂ environments. For ease, wet grinding using soluble oil, MQL with soluble oil grinding, and grinding with LN₂ as grinding medium have been named as wet, MQL, and LN₂ grinding, respectively. For wet and MQL experiments, a mixture of water-based mineral oil (Cool Edge SL, Castor oil) in 1:30 ratio has been used. The mixture of water and oil provides both excellent cooling and lubricating properties as required in the grinding process. The effectiveness of MQL primarily depends on the atomisation of the liquid droplets which primarily depends on the MQL parameters. Taking this into consideration, optimised MQL parameters as obtained by Sinha et al.³⁶ have been used. The optimised MQL parameters, grinding and dressing parameters are given in Table 3. In LN₂ grinding, liquid nitrogen jet is injected into the grinding zone through the nozzle orifice of 1 mm diameter, as schematically shown in Figure 2. Self-pressurised cryogenic Dewar (CLPB160LN2; Cryofab, USA) of filling capacity 160 L has been used to store the liquid nitrogen.

Table 3.

Experimental parameters in grinding experiments.

MQL flow rate	150 mL/h
Air pressure	8 bar
Nozzle position	Intermediate angular position
Wheel speed (V_c)	10, 15, and 20 m/s
Table speed (V_w)	3, 9, and 15 m/min
Depth of cut (d)	10, 15, and 20 µm
Dressing lead	250 mm/min
Wheel speed while dressing	18 m/s
Dressing depth	50 μm through five passes

Figure 2.

Image of the in-house developed experimental setup for (a) MQL grinding and (b) LN₂ grinding.

Details of grinding forces, surface roughness, SEM, EDS, and AFM analysis

Normal and tangential grinding forces have been measured online using Kistler 9257B piezoelectric dynamometer attached to Kistler 5070A multichannel charge amplifier. Talysurf intra surface profilometer (Taylor Hobson Limited, England) has been used to measure R_a of the ground surfaces. For R_a measurement, 0.25 mm sampling length has been chosen as per ISO 4288 (1966) standard.³⁷ The surface roughness means of five replications taken across the grinding direction have been reported in the article.

For surface characterisations, specimen of size 8 mm × 8 mm × 5 mm is cut from the ground samples under dry, wet, MQL, and LN₂ environments with wheel speed of 20 m/s, table speed of 9 m/min, and 15 μm depth of cut. These samples have been cut carefully from the ground workpieces by wire electrical discharge machining (WEDM) to avoid any surface damage. SEM tests of the ground surfaces, grinding wheel, and the grinding chips obtained after dry, wet, MQL, and LN₂ environments have been carried out using ZEISS EVO 50 Series scanning electron microscope. Since the grinding wheel is made of alumina ceramic, before SEM or EDS test, the wheel surface needs to be coated. The pieces of the wheel have been sputter coated with 10 nm coating thickness using 99% pure gold target (DC Sputter Coater, Q150RS, England). 25 mA sputter current, 5 mbar pressure, 5 min deposition time, and argon as sputtering gas have been maintained while coating the alumina wheel pieces. The three-dimensional surface topography of the ground surfaces has been examined through AFM (Bruker INNOVAA2, USA) installed with Nanoscope 8.15 software. The AFM instrument scans 50 μm × 50 μm area with a scanning rate of 256 points along a line through raster scanning. The acquired topography can be helpful in understanding the surface characteristics in a better way compared to a two-dimensional surface topography.³⁷

ANOVA

In grindability studies, it is believed that tangential force (F_t) and surface roughness are the two most important indices which are commonly used to establish the grinding behaviour of any material. Therefore, in the present work, ANOVA of F_t and R_a has been performed under all grinding environments using Design Expert 7 software. For making the work more informative, few critical parameters such as the degree of freedom (DOF), and F value of each factor and their first interaction have been reported. Then, from F distribution table data, the significance of these is checked at 5% significance level. The quadratic regression models have been developed after ignoring the insignificant terms in terms of coded variables for F_t and R_a. The coefficient of determination (R²), which reflects the degree of variability in the experimental data accounted by the regression model, has been calculated and reported. The value of R² lies in between 0 and 1.³⁸ To perform the adequacy testing of the quadratic model, adjusted R² values for each regression model has been calculated and has been reported.

Results and discussions

The measurements of grinding forces and surface roughness are important aspects in the grindability study of an engineering material. The forces generated in grinding significantly contribute to the quality of the finished products. Average surface roughness gives a valuable insight into the surface quality of the ground product. The mean values of grinding forces and R_a have been obtained from the main effects plot using the MINITAB 14 software using experimental data. For F_t and R_a under all grinding conditions, ANOVA table has been presented. Subsequently, a quadratic model using the significant factors and their first-order interaction after the F test has been used to predict the responses. Later, the samples of ground surfaces obtained after grinding under different conditions have been characterised through SEM, EDS, and AFM studies. Furthermore, grinding chips collected under different grinding environments have been characterised by SEM studies.

Variation in tangential forces (F_t)

The magnitude of F_t acting in grinding direction is the key indicator of the degree of difficulty associated with the grinding of a material under prescribed grinding conditions. Moreover, it has the vital role in the estimation of specific energy consumption in grinding. High F_t causes high heat generation and thermal damage to the work surface. From Figure 3, it can be noted that F_t decreases with increase in wheel speed, whereas increases with increase in table speed as well as the depth of cut as expected. Moreover, it is observed that F_t in MQL and LN₂ grinding are smaller than those obtained in dry and wet grinding; however, in LN₂ grinding, it has been found to be the least. In dry grinding, the values of F_t are on the higher side, which is due to the formation of wear flat over alumina grits (refer Figure 7(a)) due to high heat generation and accumulation in the grinding zone. Furthermore, these wear flattened grits are rubbed against the workpiece surface instead of performing the cutting action, leading to decreased surface roughness (see Figure 5(a)). Rubbing rather than cutting increases the friction and consequently further raises the temperature at the grinding zone.

Figure 3.

Variation in tangential forces with wheel speed, table speed, and depth of cut.

However, in other modes of grinding, relatively lower F_t has been found mainly because of cooling and lubrication provided by the cutting mediums. The percentage reduction in F_t concerning dry grinding in wet, MQL, and LN₂ grinding are 10%, 14%, and 22%, respectively. In LN₂ grinding, this decrease in F_t is due to excellent lubricity between the mating surfaces due to the formation of the lubricating film along the wheel periphery, which further assist in improving the slipping of abrasive grits over the work surface.³⁵ Formation of MQL lubricating film between the mating surfaces could be the reason behind the reduction in F_t in MQL grinding.³³ Smaller F_t further aids in retaining the sharpness of grains for longer duration resulting in lesser heat generation. At lower temperatures under MQL and LN₂ grinding, the grains retain their cutting ability for a longer spell and, therefore, generate lower tangential forces and eventually result in increased wheel life.^35,39 From this, it can be established that MQL and LN₂ grinding condition leads to lesser power consumption resulting in lower heat generation which leads to a favourable situation for grinding of IN718.

Variation in normal forces (F_n)

Normal force variations in dry, wet, MQL, and LN₂ with wheel speed, table speed, and depth of cut are shown in Figure 4. From Figure 4, it can be comprehended that F_n shows declining trend with an increase in wheel speed. Moreover, it can also be found that F_n is slightly lesser in MQL grinding compared to wet grinding, whereas in LN₂ grinding, they are always smaller. Smaller F_n signifies the ease in indentation which eventually results in easy grinding. The application of soluble oil in MQL mode and LN₂ jets significantly reduces normal force component thereby ensuring the retention of grit sharpness for a longer duration due to a reduction in grinding zone temperature resulting in lesser wheel wear.³⁵ Overall percentage reduction in F_n by 10%, 15%, and 20% in the case of wet, MQL, and LN₂ grinding has been found. Moreover, Figure 4 shows an increase in F_n with an increase in table speed and depth of cut, which well agrees with the grinding basics. In dry and wet grinding, F_n is much higher compared to MQL and LN₂ grinding that can cause erratic part tolerance and self-induced vibration, known as chatter.

Figure 4.

Variation in normal forces with wheel speed, table speed, and depth of cut.

Variation in centre line average surface roughness (R_a)

The change in R_a with wheel speed, table speed, and depth of cut has been shown in Figure 5(a). Here, it can be noted that R_a decreases with increase in wheel speed, whereas almost reverse trend has been observed with increase in table speed and infeed. This variation is in concurrence with the basic grinding principles. In dry grinding, R_a has been found to be the least compared to the other grinding conditions. It can be attributed to the severe wear flattening of alumina grits, which is apparent by the formation of a smooth, glazed, and shiny wheel surface. Glazing is due to relatively higher grinding forces which result in more blunting of the grits as can be noted in Figure 7 causing a high heat generation. In MQL grinding, larger R_a has been found in all cases as compared to dry, wet, and LN₂ grinding. This may be due to the possible retention of sharp facets by the cutting abrasive grits for a greater period of time. The smaller grinding forces in MQL also supports this belief. Breakage of MQL film leading to partial direct contact between the mating surfaces may be another reason for observing higher R_a in MQL grinding. This argument appears to be in tune with the obtained result of slight surface oxidation in MQL grinding as will be discussed in the upcoming section. The R_a in MQL grinding is on the higher side, but the values are in the tune of 0.1 and 0.15 µm higher as compared to the LN₂ and dry grinding. In the industrial practices, this much variation does not raise any serious concern. In LN₂ grinding, R_a is much lower than that observed in MQL grinding, which indicates the improved lubricity effect due to the formation of a thin lubricating layer along the grinding wheel periphery. The onset of wear flats over the grits due to the higher bond strength at a cryogenic temperature in LN₂ grinding may be another reason for the lower R_a (refer Figure 7) values. The lower temperature in LN₂ grinding does not allow easy dislodgement of the worn-out grits, and hence, the grinding with such grits leads to initiation of wear flattening.¹⁸ Figure 5(b) presents the main effects and interaction effects plot for R_a in the case of dry, wet, MQL, and cryogenic grinding conditions. These plots are helpful in understanding the effect of the individual factors as well as their interactions on the surface roughness.

Figure 5.

(a) Variation in average surface roughness with wheel speed, table speed, and depth of cut; (b) main effects and interaction plots for mean surface roughness in dry, wet, MQL, and LN₂.

Wheel surface and ground surface topography investigations

Figure 6 shows the surface topography of alumina wheel after completion of grinding experiments under dry, wet, and LN₂ grinding. From this figure, severe deposition of grinding chips on wheel surface in dry grinding can be seen, whereas in MQL grinding, the deposition is relatively mild. In the case of LN₂ grinding, almost negligible amount of chip deposition has been obtained, which is possibly due to better flush away of the chips by the pressurised LN₂ jet. Figure 7 shows the SEM images of the grinding wheel showing the excessive wear flat formation in the case of dry grinding. Consequently, the wear land over the grits leads to the wheel dulling and results in increased level of rubbing consuming more energy in dry grinding. The wear flat over the grits consequently results in wheel glazing with the increase in number of grinding passes, leading to relatively smoother ground surface in dry grinding. In LN₂ grinding, such deposition and redeposition are almost negligible, and therefore, orderly splintering of grits occurred resulting in the retention of their sharpness. The sharper grits, in consequence, resulted in relatively higher roughness. In MQL grinding with soluble oil, a minute amount of the debris deposition has been found. Figure 8 shows the EDS analysis of the wheel surfaces after grinding under dry, wet, MQL, and LN₂ conditions. The likelihood of the highest flash temperature in dry grinding is also supported by the maximum percentage of oxygen (around 40.7%), whereas in MQL, wet, and LN₂ grinding, the corresponding values are 33.9%, 30.4%, and 26.6%, respectively (refer Figure 8).

Figure 6.

Grinding wheel images after the grinding experiments under (a) dry, (b) MQL, and (c) LN₂ grinding.

Figure 7.

SEM micrographs of the wheel surfaces obtained after grinding experiments under (a) dry, (b) wet, (c) MQL, and (c) LN₂ grinding conditions.

Figure 8.

EDS analysis of alumina wheel surface post grinding under (a) dry, (b) wet, (c) MQL, and (d) LN₂ environments.

AFM surface topographic images of the work surface ground under dry, MQL, and LN₂ conditions are presented in Figure 9. The dimensions of the AFM images of the ground surfaces obtained after dry, MQL, and LN₂ conditions are 50 μm × 50 μm × 0.807 μm, 50 μm × 50 μm × 1.3 μm, and 50 μm × 50 μm × 0.974 μm, respectively. The average height of the scanned area over the ground surfaces obtained under dry, LN₂, and MQL grinding is 0.807, 0.974, and 1.3 μm, respectively, which shows an increasing trend. The least average height is found in dry grinding and the highest in MQL grinding which also supports the reasoning of intense wheel flat in dry grinding and grit micro-fracturing in MQL grinding. The R_a values obtained from AFM for dry, MQL, and LN₂ grinding environments are 158, 316, and 259 nm, respectively, whereas the height of highest peak (R_max) values are 1727, 2126, and 2061 nm, respectively. From these images, the relatively smoother surface has been observed in dry grinding, which is primarily because of intense rubbing due to wear flats on the grinding wheel. Moreover, the deposition of debris as shown in Figure 7 between the grit interspacing and over the grits makes the wheel surface shiny and glazed.

Figure 9.

AFM image of the ground surface obtained after grinding under (a) dry, (b) MQL, and (c) LN₂ environments.

ANOVA analysis of F_t and R_a

ANOVA of the two most significant grinding attributes F_t and R_a has been performed using Design Expert 7. With the individual grinding parameters’ effect, first-order interaction between the grinding parameters has been established through F test at 5% significance level. The calculated F values for F_t and R_a have been compared to an F value taken from the F distribution table for all the grinding environments as depicted in Table 4. Based on the relative F values, significant terms have been identified and marked with a star (*) mark. Then, after removing the insignificant terms, second-order multiple regression models in coded form have been fitted for both responses as given below in the form of equations. Furthermore, to ensure the adequacy of these models, the predicted and actual values have been compared, and the relations between the predicted and the actual values for R_a and F_t have been cited in Figure 10. In addition to the R² value, the adjusted R² value of these responses in all situations has been calculated using the software package and has been shown in Table 5. The adjusted R² values of F_t and R_a have been found to be fairly close to R² values. This closeness indicates that the adopted quadratic regression model is sufficient to capture the experimental data variations. Additionally, it also provides information that by further increasing the degree of the regression model, there will not be any improvement in the regression fit as obtained. The errors are also found to be randomly distributed not showing any distinct trend which is a prerequisite condition for checking the adequacy of the models³⁸

\begin{array}{l} F_{t} (Dry) = 36.57 - 6.12 \times V_{c} + 8.20 \times V_{w} + 8.81 \times d \\ - 3.03 \times V_{c} \times V_{w} - 5.59 \times V_{c}^{2} + 4.90 \times V_{w}^{2} \end{array}

(1)

\begin{array}{l} F_{t} (Wet) = 32.90 - 5.80 \times V_{c} + 4.02 \times V_{w} \\ + 9.79 \times d - 1.24 \times V_{c} \times V_{w} \\ - 1.55 \times V_{c} \times d + 1.94 \times V_{c}^{2} - 2.91 \times V_{w}^{2} \end{array}

(2)

\begin{array}{l} F_{t} (MQL) = 32.60 - 5.24 \times V_{c} + 4.39 \times V_{w} \\ + 9.42 \times d - 2.95 \times V_{c} \times d - 2.69 \times V_{w}^{2} \end{array}

(3)

\begin{array}{l} F_{t} ({LN}_{2}) = 28.43 - 5.69 \times V_{c} + 3.75 \times V_{w} \\ + 8.72 \times d - 3.69 \times V_{c} \times d \end{array}

(4)

\begin{array}{l} R_{a} (Dry) = 0.36 - 0.018 \times V_{c} + 0.016 \times V_{w} + 0.020 \times d \\ + 7 E - 003 \times V_{c} \times d - 0.010 \times V_{c}^{2} + 0.013 \times d^{2} \end{array}

(5)

\begin{array}{l} R_{a} (Wet) = 0.43 - 0.037 \times V_{c} + 0.028 \times V_{w} \\ + 0.024 \times d + 0.027 \times V_{c} \times V_{w} + 0.011 \times V_{w}^{2} \end{array}

(6)

\begin{array}{l} R_{a} (MQL) = 0.52 - 0.041 \times V_{c} + 0.015 \times V_{w} + 0.023 \times d \\ + 0.018 \times V_{c} \times d + 7.25 E - 003 \times V_{w} \times d + 0.013 \times d^{2} \end{array}

(7)

R_{a} ({LN}_{2}) = 0.42 - 0.019 \times V_{c} + 0.046 \times V_{w} + 0.038 \times d

(8)

Table 4.

Comparison of F values for F_t and R_a for different grinding conditions.

Sources of variation	Degree of freedom	Dry grinding		Wet grinding		MQL grinding		LN₂ grinding
		F_t	R_a	F_t	R_a	F_t	R_a	F_t	R_a
V_c	1	36.39*	61.67*	213.93*	52.82*	100.92*	196.25*	60.44*	21*
V_w	1	65.35*	47.90*	102.79*	30.46*	70.92*	28.74*	26.25*	128.49*
d	1	75.50*	77.95*	609.74*	21.57*	326.18*	62.20*	142.18*	84.47*
V_c·V_w	1	7.12*	0.80	7.80*	21.41*	2.75	3.74	0.028	0.81
V_c·d	1	0.078	7.72*	12.25*	0.023	25.63*	29.31*	20.39*	7.428E–004
V_w·d	1	0.16	0.35	0.19	2.66	1.36	5.03*	3.374E–003	0.036
$V_{c}^{2}$	1	5.86*	5.07*	5.16*	2.84	0.89	0.26	0.034	0.065
$V_{w}^{2}$	1	9.03*	0.076	17.10*	5.24*	5.43*	1.11	0.54	0.33
d ²	1	1.58	9.60*	0.59	0.096	2.45	10.29*	0.95	4.085E–003
Residual	10
Total	19	F_{0.05, 1, 10} = 4.9646 (from F distribution table)³⁸

Figure 10.

Comparison between predicted and actual values of surface roughness and tangential forces.

Table 5.

R² and adjusted R² values for F_t and R_a.

		Dry grinding	Wet grinding	MQL grinding	LN₂ grinding
Tangential force (F_t)	R ²	0.9429	0.9889	0.9698	0.9431
	Adjusted R²	0.9166	0.9825	0.9591	0.9279
Mean surface roughness (R_a)	R ²	0.9484	0.8858	0.9544	0.9539
	Adjusted R²	0.9245	0.8450	0.9344	0.9453

SEM and EDS test of ground surfaces

Figures 11 and 12 present the SEM and EDS images of the ground surfaces generated under different grinding environments, respectively. Distinct thermal cracks on the dry ground surface can be seen in Figure 11(a). These thermal cracks are due to high heat generation and more importantly, heat accumulation over the ground surface as a result of the lesser thermal conductivity of the work surface. From Figure 11(d), it is clear that cryogenic grinding reflects the best ground surface, whereas in dry grinding, surface cracks and severe redeposition can be seen. However, in wet and MQL grinding, the redeposition is relatively milder which can be attributed to the relatively lesser heat generation compared to the dry grinding. Figure 12 depicts the EDS analysis of the ground surface under dry, wet, MQL, and LN₂ environments. The elemental composition of these test samples has been given in Table 6. The weight percentage of oxygen in dry, wet, MQL, and LN₂ ground surface is 10.858, 4.192, 5.202, and 2.989, respectively, which indicates severe oxidation in the case of dry grinding compared to the other three. The decrement in Ni percentage from the LN₂ ground surface: 42.062, MQL: 39.498, wet: 40.447, and dry: 31.388, respectively, also support the formation of nickel compounds primarily due to more heat generation in dry grinding. From this, it can be opined that minute oxidation has been found in the case of MQL grinding, whereas LN₂ grinding generates the surface free from surface oxidation. This is due to the rapid heat extraction from the grinding zone by the LN₂ jet leading to completely unoxidised ground surface. In wet grinding, the level of surface oxidation is slightly lesser than the surface oxidation obtained in MQL grinding. This improvement in wet grinding is at the high cost of the coolants used and possesses a threat to the operator and the environment. MQL grinding generates the ground surface having oxygen content in the level 5.202% which is marginally higher than the level of oxygen (4.192%) in wet grinding. This insignificant decrease in surface oxidation using litres of cutting fluid does not seem a viable and economical solution. Therefore, MQL grinding can be considered as a realistic alternative considering the cost and quality aspects of the ground component.

Figure 11.

Ground surface SEM micrographs under (a) dry, (b) wet, (c) MQL, and (c) LN₂ environments.

Figure 12.

Ground surface EDS micrographs under (a) dry, (b) wet, (c) MQL, and (d) LN₂ environments.

Table 6.

Elemental composition of ground surfaces obtained after different grinding conditions.

Elements	Weight (%)
	Dry grinding	Wet grinding	MQL grinding	LN₂ grinding
Nickel	31.388	40.447	39.498	42.062
Oxygen	10.858	4.192	5.202	2.989
Carbon	22.578	17.993	15.282	16.583
Iron	13.276	14.236	15.494	16.396
Chromium	12.031	15.164	14.114	14.057
Niobium	5.336	3.861	5.382	3.779
Molybdenum	3.287	2.523	2.705	2.668
Titanium	0.800	0.862	1.636	0.753
Aluminium	0.445	0.560	0.687	0.714

SEM characterisation of grinding chips

The SEM micrographs of the grinding chips collected on carbon tape under different grinding environments are depicted in Figure 13. From Figure 13, it is clear that in dry grinding, almost complete curling of chips (i.e. almost circular chip) has been observed, which proves that high heat generation and severe oxidation occurred. The curling of chips is more common, while grinding with conventional wheel such as alumina wheel possesses small thermal conductivity and therefore, more heat is flown to the grinding chips.⁴⁰ In the case of wet and MQL grinding, chip sizes are relatively smaller caused by chip breakage due to relatively high cooling effect compared to dry grinding. However, in LN₂ grinding, almost C-type chip has been found which is due to the sudden and intense cooling of chips leading to its breakage from the middle (refer Figure 13(d)). In wet grinding, the chip size is random consisting of large as well as C-type chip, which hints towards the reduced level of cooling compared to cryogenic grinding. In MQL grinding, broken chips can be seen (refer Figure 13(c)) that explicitly points towards the reasonable high cooling effects. These chip morphologies also support the effectiveness of MQL grinding. However, the LN₂ grinding seems to be better in grinding of high-strength IN718.

Figure 13.

SEM micrographs of chips obtained under (a) dry, (b) wet, (c) MQL, and (d) LN₂ grinding.

Conclusion

The present work is an attempt in the direction of grindability improvement of IN718 superalloy using ecological grinding environments such as MQL with soluble oil and LN₂. From this work, the key conclusions can be summarised as follows:

In dry grinding, both tangential and normal forces have been found to be always larger primarily because of profound wear flattening of the grits. Thermal cracks over the ground surface, lower average surface roughness, and almost circular chip justify this likelihood in dry grinding.

In MQL grinding, lower tangential forces have been observed, which indicate favourable grinding condition as compared to dry grinding. From SEM micrographs and EDS analysis, it is concluded that MQL grinding resulted in minor surface oxidation. Marginally, higher mean surface roughness has been found in the MQL ground surface which may be due to better retention of the sharp cutting edges of the grinding grits.

MQL grinding can be a possible alternative to IN718 grinding. MQL grinding will not only enhance the grinding process effectiveness but will also significantly reduce the manufacturing costs by lowering the cutting fluid consumption and other associated costs.

The LN₂ environment provides a significant reduction in grinding forces and grinding zone temperature, which consequently promotes favourable grit–work surface interactions. Moreover, the inert nature of LN₂ provides cooling and lubrication without damaging the ground surface. This argument is also supported by the lowest oxygen percentage found in the case of a ground surface and wheel surface obtained under LN₂ grinding through EDS test results.

SEM images of the ground wheel surface reveal the excessive debris deposition over the wear flat of alumina abrasives in dry grinding, whereas grit wear flattening is almost invisible in LN₂ grinding.

Wheel glazing due to wear flats in dry grinding resulted in smoother ground surface compared to MQL and LN₂ grinding. The three-dimensional AFM surface analysis well supports this postulate through the topographic images obtained.

The quadratic regression model has been fitted to predict the tangential forces and mean surface roughness. The presented models have been found to be adequate in capturing the variance in the experimental data.

From the EDS analysis of the ground surface obtained under dry, MQL, and LN₂ conditions, it is also evident that ground surface has undergone severe oxidation under dry grinding reflected by the highest value of oxygen weight percentage compared to MQL and LN₂ grinding.

Footnotes

Acknowledgements

The authors thank the anonymous reviewers for their comments which led to improvements in this article.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID ID

Manoj Kumar Sinha

References

Cheng

Machining dynamics: fundamentals, applications and practices. Berlin: Springer, 2008.

Malkin

Guo

Grinding technology–theory and applications of machining with abrasives. 2nd ed. New York: Industrial Press, 2008.

Hadad

Hadi

An investigation on surface grinding of hardened stainless steel S34700 and aluminum alloy AA6061 using minimum quantity of lubrication (MQL) technique. Int J Adv Manuf Tech 2013; 68(9–12): 2145–2158.

Sieniawski

Nadolny

The effect upon grinding fluid demand and workpiece quality when an innovative zonal centrifugal provision method is implemented in the surface grinding of steel CrV12. J Clean Prod 2015; 113: 960–972.

Agarwal

Optimizing machining parameters to combine high productivity with high surface integrity in grinding silicon carbide ceramics. Ceram Int 2016; 42(5): 6244–6262.

Shao

Fergani

Ding

et al . Experimental investigation of residual stress in minimum quantity lubrication grinding of AISI 1018 steel. J Manuf Sci E: T ASME 2016; 138(1): 011009.

Dixit

Sarma

Davim

JP.

Environmentally friendly machining. New York: Springer, 2012.

Ramesh

Yeo

Zhong

et al . Ecological grinding with chilled air as coolant. Proc IMechE, Part B: J Engineering Manufacture 2003; 217(3): 409–419.

Weinert

Inasaki

Sutherland

et al . Dry machining and minimum quantity lubrication. CIRP Ann Manuf Techn 2004; 53(2): 511–537.

10.

Debnath

Reddy

QS.

Environmental friendly cutting fluids and cooling techniques in machining: a review. J Clean Prod 2014; 83(0): 33–47.

11.

Nascimento

WRD

Yamamoto

Mello

HJD

et al . A study on the viability of minimum quantity lubrication with water in grinding of ceramics using a hybrid-bonded diamond wheel. Proc IMechE, Part B: J Engineering Manufacture 2015: 1–9.

12.

Chetan Ghosh

Rao

PV.

Application of sustainable techniques in metal cutting for enhanced machinability: a review. J Clean Prod 2015; 100: 17–34.

13.

Sanchez

Pombo

Alberdi

et al . Machining evaluation of a hybrid MQL-CO2 grinding technology. J Clean Prod 2010; 18(18): 1840–1849.

14.

Emami

Sadeghi

Sarhan

AAD

et al . Investigating the minimum quantity lubrication in grinding of Al2O3 engineering ceramic. J Clean Prod 2014; 66: 632–643.

15.

Tawakoli

Hadad

Sadeghi

et al . Minimum quantity lubrication in grinding: effects of abrasive and coolant–lubricant types. J Clean Prod 2011; 19: 2088–2099.

16.

Setti

Ghosh

Paruchuri

. Influence of nanofluid application on wheel wear, coefficient of friction and redeposition phenomenon in surface grinding of Ti-6Al-4V. Proc IMechE, Part B: J Engineering Manufacture 2016: 1–13.

17.

Kumar

Ghosh

Aravindan

Grinding performance improvement of silicon nitride ceramics by utilizing nanofluids. Ceram Int 2017; 43(16): 13411–13421.

18.

Reddy

Ghosh

Some critical issues in cryo-grinding by a vitrified bonded alumina wheel using liquid nitrogen jet. J Mater Process Tech 2016; 229: 329–337.

19.

Thakur

Gangopadhyay

State-of-the-art in surface integrity in machining of nickel-based super alloys. Int J Mach Tool Manu 2016; 100: 25–54.

20.

Zhang

Zhao

Aerospace materials handbook. Boca Raton, USA: CRC Press, 2012.

21.

Sinha

Ghosh

Paruchuri

VR.

Modelling of specific grinding energy for Inconel 718 superalloy. Proc IMechE, Part B: J Engineering Manufacture 2019; 233(2): 443–460.

22.

Chetan Ghosh

Rao

PV.

Environment friendly machining of Ni–Cr–Co based super alloy using different sustainable techniques. Mater Manuf Process 2016; 31(7): 852–859.

23.

Tobler

RL.

Low temperature effects on the fracture behaviour of a nickel base superalloy. Cryogenics 1976; 16(11): 669–674.

24.

Marques

Narala

SKR

Machado

et al . Performance assessment of MQSL: Minimum quantity solid lubricant during turning of Inconel 718. Proc IMechE, Part B: J Engineering Manufacture 2017; 231(7): 1144–1159.

25.

Sinha

Madarkar

Ghosh

et al . Application of eco-friendly nanofluids during grinding of Inconel 718 through small quantity lubrication. J Clean Prod 2017; 141: 1359–1375.

26.

Behera

Alemayehu

Ghosh

et al . A comparative study of recent lubri-coolant strategies for turning of Ni-based superalloy. J Manuf Process 2017; 30: 541–552.

27.

Jia

et al . Influence of cutting tool geometrical parameters on tool wear in high-speed milling of Inconel 718 curved surface. Proc IMechE, Part B: J Engineering Manufacture 2019; 233(1): 18–30.

28.

Xue

Lijun

Hui

et al . Superplastic properties of Inconel 718. J Mater Process Tech 2003; 137(1–3): 17–20.

29.

Schirra

Viens

. Metallurgical factors influencing the machinability of Inconel 718. Superalloys 718, 625, 706 and various derivatives. Pittsburgh, PA: The Minerals, Metals &Materials Society, 1994: 827–838.

30.

Sieniawski

Nadolny

Experimental study into the grinding force in surface grinding of steel CrV12 utilizing a zonal centrifugal coolant provision system. Proc IMechE, Part B: J Engineering Manufacture 2018; 232(3): 394–403.

31.

Pavan

Gopal

Amrita

et al . Experimental investigation of graphene nanoplatelets–based minimum quantity lubrication in grinding Inconel 718. Proc IMechE, Part B: J Engineering Manufacture 2019; 233(2): 400–410.

32.

Tawakoli

Hadad

Sadeghi

MH.

Influence of oil mist parameters on minimum quantity lubrication – MQL grinding process. Int J Mach Tool Manu 2010; 50: 521–531.

33.

Barczak

Batako

ADL

. Application of minimum quantity lubrication in grinding. Mater Manuf Process 2011; 27: 406–411.

34.

García

Méresse

Pombo

et al . Role of frozen lubricant film on tribological behaviour and wear mechanisms in grinding. Int J Adv Manuf Tech 2016; 82(5–8): 1017–1027.

35.

Setti

Yadav

Ghosh

Grindability improvement of Ti-6Al-4V using cryogenic cooling. Proc IMechE, Part B: J Engineering Manufacture 2014; 228(9): 1131–1137.

36.

Sinha

Setti

Ghosh

et al . An alternate method for optimization of minimum quantity lubrication (MQL) parameters in surface grinding. Int J Mach Mach Mater 2016; 18(5–6): 586–605.

37.

Griffiths

Manufacturing surface technology: surface integrity and functional performance. 1st ed. London: Penton Press, 2001.

38.

Montgomery

DC.

Design and analysis of experiments. Hoboken, NJ: John Wiley & Sons, 2008.

39.

Paul

Bandyopadhyay

Chattopadhyay

Effects of cryo-cooling in grinding steels. J Mater Process Tech 1993; 37(1): 791–800.

40.

Rajmohan

Radhakrishnan

A study on the thermal aspects of chips in grinding. Int J Mach Tool Manu 1992; 32(4): 563–569.

Some investigations in grindability improvement of Inconel 718 under ecological grinding

Abstract

Keywords

Introduction

Materials and methods

Deciding the experimental runs

Work material, grinding wheel, grinding environments, and grinding kinematics

Details of grinding forces, surface roughness, SEM, EDS, and AFM analysis

ANOVA

Results and discussions

Variation in tangential forces (Ft)

Variation in normal forces (Fn)

Variation in centre line average surface roughness (Ra)

Wheel surface and ground surface topography investigations

ANOVA analysis of Ft and Ra

SEM and EDS test of ground surfaces

SEM characterisation of grinding chips

Conclusion

Footnotes

Acknowledgements

Declaration of conflicting interests

Funding

ORCID ID

References

Variation in tangential forces (F_t)

Variation in normal forces (F_n)

Variation in centre line average surface roughness (R_a)

ANOVA analysis of F_t and R_a