1 Introduction

Aluminum (Al)-based alloys have become the best alternative to conventional high strength materials owing to their good strength to weight ratio, better mechanical and chemical properties, i.e., resistance to high temperature, high stiffness [1, 2]. Aluminum-based alloys are mostly used in aerospace (aircraft fittings, aircraft structural components), marine applications, and automotive industry (shafts, truck wheels) as a shaft material. The usage of aluminum alloys in the manufacturing industry has noticeably risen in the past few years. Low weight and high strength combined in aluminum alloys have made them a preferable choice for the aerospace and automotive industry. This key feature of aluminum alloys makes it a fair substitute for steel and cast iron for the fabrication of different parts in the aerospace and automotive industry. Moreover, the presence of copper in the Al 2024 alloy makes it heat-treatable so its machinability can be improved [1, 3, 4]. We have chosen Al 2024 alloy as our workpiece to aid the said industry in the machining of this special material.

In machining of aluminum alloys, the use of cutting fluids will give less tool wear and better surface roughness [5, 6]. But because of the environmental risks and economical costs linked with the cutting fluids, dry machining is being preferred nowadays in the manufacturing industry. Dry machining has got attention in industry as compared to wet machining due to environmental hazardous and cost coupled with the usage of cutting fluids [7]. The cost of cutting fluids is 16–20% of total production cost and this cost is in addition to the cost of equipment of coolant spray, disposal of cutting fluid, and its hostile effects on environment and operator [8]. Thus, there is a growing need for dry machining in the industry for environmental safety and waste reduction [9].

Tool wear and workpiece surface roughness are two main important factors on which the machining performance is evaluated [10, 11]. Workpiece surface roughness and tool wear greatly affects the service and performance of finished product and production cost. Conventionally, the utilization of cutting fluids helped to achieve the less tool wear and better workpiece surface roughness [12]. However, with the paradigm shift from wet to dry machining, problems to attain less tool wear and better workpiece surface roughness have increased. The selection of suitable machining parameters is becoming extremely crucial to achieve the best machining performance [13]. Besides this, the selection of a suitable cutting tool is equally essential to get good machining performance [14]. Different types of cutting tools are employed for the machining of the aluminum alloys, but it is recommended to use coated tools in dry machining of aluminum alloys [15, 16].

The selection of coated tools based on their mechanical and adhesion strength is also necessary for the machining performance. Generally, it was thought that harder cutting tools would have better machining performance. Apart from the hardness of the tools, the adhesion strength of the coated tools contributes equally to assessing the machining performance [17,18,19]. There is no comprehensive research regarding the causal relationship between the mechanical properties, adhesion strength of the cutting tools, tool wear, and workpiece surface roughness in literature. Thus, to get maximum benefits out from dry machining, there is a need to carry out such research in which experimental investigation would be done to probe the failure mechanisms of coated tools and the causal relationship of tool wear and surface roughness with mechanical properties and adhesion strength of the cutting tools.

In this study, aluminum alloy 2024 is used in dry machining with four different cutting inserts, i.e., cemented carbide, TiN, TiAlN, and PCD coated. Effect of machining parameters on the tool wear and workpiece surface roughness was evaluated by DOE and RSM. Optimization of the machining parameters was conducted for all the four cutting inserts. Comparative performance analysis of tool wear and workpiece surface roughness has been carried out for all four cutting inserts. Moreover, the effect of mechanical properties and adhesion of cutting inserts on tool wear and workpiece surface roughness have been evaluated, and failure mechanism of coated tools has been validated by EDX analysis of the chips.

2 Experimentation and methodology

2.1 Material and experimental setup

A circular cross-sectional bar of aluminum alloy 2024 of 40 mm diameter was selected for experimental work. Its chemical composition, given by the supplier, which is according to ASM standards is given in Table 1 [20].

Table 1 Chemical composition of aluminum alloy 2024
Full size table

The workpieces were machined on a lathe machine (4NE-II 600) in dry conditions. The specifications of the lathe machine are given in Table 2. Pre-machining turning operation was performed to ensure uniformity of diameter along the length of the bar and to reduce the chances of error. Tailstock along with headstock was engaged to reduce vibrations which ensured error-free experimentation. After machining tools were washed in an ultrasonic bath for 15 min. The turning time for each experiment was 30 min. The turning experiments were performed using four types of inserts, i.e., TiN, TiAlN, PCD coated on cemented carbide inserts, and uncoated cemented carbide. Insert geometry is ISO designated VNMG160408 general purpose grade. Cutting inserts have the following geometrical characteristics: cutting edge length = 16 mm, insert thickness = 4 mm, nose radius = 0.8 mm, cutting edge angle = 72.5°. SVJBR 2525M16 tool handle was utilized for all the experiments.

Table 2 Specifications of the lathe machine used for machining
Full size table

2.2 Measurement of the workpiece surface roughness

Workpiece surface roughness (Ra) was evaluated by Insize Plus (ISR-1000B) portable surface roughness tester, having an accuracy of ± 3% and a resolution of 0.001 μm. The cutoff length and measuring speed of the surface roughness tester were set to 0.8 mm and 1 mm/s, respectively. Surface roughness was measured at five different places and then average of these values was recorded. In this study, workpiece surface roughness was chosen as a machining performance indicator because cutting tools have a great impact on the workpiece surface roughness [21, 22].

2.3 Measurement of tool wear

The tool wear, after machining of 30 min, was measured by CE-450 DV optical coordinated measuring machine (CMM), manufactured by CHEN WEI Precise Technology Ltd., having a maximum resolution of 700 × [23]. Five measurements were taken over the whole wear profile and each measurement was recorded three times using Quadra-Check (QC) 5000 software. The average of these values was recorded as the tool wear for all the 108 experiments.

2.4 Measurement of adhesion strength and mechanical properties of tools

The adhesion strength of the TiN, TiAlN, and PCD coated tools was determined by the scratch tester of CSM Instruments (CSM MSTX S/N: 01-2569) [24]. Rockwell-type diamond indenter of 100 μm radius was used to scratch off the coating. The applied normal load starts from 0.03 and goes up to 20 N in these experiments. Scratch length and loading and unloading rate were kept 3 mm and 9.99 N/min, respectively. The adhesion of the coating was measured from the critical loads. The critical load is the load at which complete delamination of the coating occurs and the substrate becomes apparent [25]. The equipment measures the critical load, acoustic emission, and penetration depth simultaneously. Acoustic emissions are propagated at the failure point of the coating, and the strong signal is produced which indicates the point of critical load measurement. Acoustic emission and optical micrographs of the scratch tracks were both utilized to decide the adhesion of the coating [25,26,27].

Hardness and elastic modulus of the cutting inserts were evaluated, by the standard nanoindentation method [28,29,30], with the nanohardness tester of CSM instruments (NHTX S/N: 01-2569). Berkovich diamond indenter was used for the indentation in the experiments. A maximum of 40.0 mN load was employed, and the loading and unloading rate was set to 80.0 mN/min during indentation. The results were acquired in the form of load-displacement curves. Then Oliver and Pharr method was adopted to determine the hardness and elastic modulus of the coatings [31].

2.5 EDX analysis

Aluminum alloy 2024 workpieces were machined at optimum cutting parameters for a single stroke of turning operation, for all the four cutting inserts (cemented carbide, TiN, TiAlN, and PCD coated). The chips from each experiment were collected separately, and their chemical composition was analyzed via EDX analysis [32, 33]. EDX analysis was crucial in our study as it substantiated our major findings regarding the influence of adhesion strength of coatings on the machining performance.

2.6 Design of experiment

A full factorial three-level DOE was performed in this study. The DOE is a very powerful technique of analyzing the effect of input process variables on the response variables [11, 34, 35]. To examine the impact of machining parameters on the tool wear and the surface roughness of workpiece, three most fundamental parameters, namely, cutting speed (CS), feed rate (FR), and depth of cut (DOC) were chosen for this study [36, 37]. These parameters were taken as the independent input variables while the output responses were tool wear and surface roughness. Total 33 = 27 experiments were carried out for all four types (cemented carbide, TiN, TiAlN, and PCD coated) of cutting inserts. All the machining parameters and their corresponding ranges are given in Table 3. Ranges of machining parameters were selected from the literature [38,39,40,41]. The complete detail of the methodology and experimental setups are given in Fig. 1.

Table 3 Input machining variables with their levels
Full size table
Fig. 1
figure 1

Flow chart for the methodology and experimental setups

Full size image

3 Results and discussion

3.1 Machining experiments

The results obtained from the machining of Al 2024 workpieces based on DOE for all the four cutting inserts and their responses are given in Tables 4 and 5, respectively. Tables 4 and 5 present tool wear and workpiece surface roughness values against all the 27 experiments of four cutting inserts. Radar charts are constructed from Tables 4 and 5 to get a comprehensive visual comparison of the machining performance of the four cutting inserts. The radar charts are an effective means of representing multivariant data in a two-dimensional graph [42].

Table 4 Design of experiments and the tool wear of all four types of cutting inserts
Full size table
Table 5 Design of experiments and workpiece surface roughness for all four types of cutting inserts
Full size table

Figure 2 compares the tool wear of the four cutting inserts for all the 27 experiments. PCD coated inserts exhibit maximum tool wear and TiAlN coated inserts display minimum tool wear for all the 27 experiment tools (minimum tool wear indicates better machining performance). The order of better machining performance regarding tool wear is as follows: TiAlN coated > uncoated cemented carbide (CC) > TiN coated > PCD coated.

Fig. 2
figure 2

Radar graph for tool wear of all the cutting tools against all 27 experiments

Full size image

Similarly, Fig. 3 shows the radar chart for workpiece surface roughness for all the 27 experiments. Like the results of tool wear, PCD once again performed worst for workpiece surface roughness. TiN coated inserts displayed excellent results of workpiece surface roughness as compared to the other cutting tools (minimum surface roughness indicates better machining performance). The order of better machining performance regarding workpiece surface roughness is as follows: TiN coated > uncoated CC > TiAlN coated > PCD coated. It is important to note that a consistent pattern emerges throughout the 27 experiments for both the tool wear and surface roughness.

Fig. 3
figure 3

Radar graph for surface roughness of all the cutting tools against all 27 experiments

Full size image

3.2 Response surface methodology

Response surface methodology was employed to examine the effect of machining parameters on tool wear and workpiece surface roughness [43, 44]. RSM is a powerful method that can explore the relationship between input variables and one or more output response variables [45]. In this study, DOE-based RSM was performed on Minitab® by considering linear, quadratic, and two-way interactions. Table 6 gives the coded coefficients of the regression model for CC, TiN, TiAlN, and PCD coated inserts. Single terms such as DOC, CS, and FR are linear terms while terms such as DOC*CS and DOC*DOC represent two-way and quadratic interactions, respectively. The term “Coef” represents regression coefficients, and these coefficients tell us about the size and direction of the relationship among the input and output variables. Also, these coefficients are the constant terms that are multiplied in a regression equation. The size of the coefficients gives us information about the amount of effect a term has on the response variables while the sign shows the direction of the relationship [43]. The term “SE Coef” represents the standard error of the coefficients. It shows the precision to measure the coefficients; the smaller the error, the more precise we can determine the coefficients. The “p” values show the significance of the term in the regression equation based on a 95% confidence interval. The terms which have a “p” value less than 0.05 are significant and have a strong effect on the desired responses [46, 47].

Table 6 Coded coefficients for tool wear and workpiece surface roughness of all the four cutting inserts
Full size table

Furthermore, the results of the response surface methodology are given in Table 7. The regression model for tool wear of TiAlN coated inserts presents the highest R2 value (99.33%) and thus shows the example of an excellent regression model, while surface roughness model for the cemented carbide regression model represents an excellent fit among other tools with 97.18% R2 value. R2 is called the coefficient of determination and its value is always between 0 and 100%; higher R2 value means better fit [43, 48]. In our study, the R2 values of all the response surface regression models are greater than 0.95 which demonstrates an excellent fit. Moreover, R2 values are closer to the R2 adjusted values which prove the fitness of our models [49].

Table 7 Summary of response surface methodology
Full size table

3.3 Statistical analysis of the responses on machining parameters

Figure 4 illustrates the main effect plots of tool wear of all four types of inserts. These graphs indicate that tool wear is considerably influenced by the depth of cut for all the cutting inserts. Depth of cut has a positive linear effect on the tool wear for all the cutting inserts: as the depth of cut increases, the corresponding tool wear also increases. Since the increase in depth of cut increases the tool-workpiece contact area, the outcome is the increase in the tool wear [50]. Similarly, cutting speed has a linear positive effect on tool wear after the depth of cut for all the inserts. Hence, it is imperative to note that the tool wear constantly increased as the depth of cut and cutting speed increased. However, there is a nonlinear positive effect of feed rate on tool wear of all the cutting inserts. Tool wear increases with the increase in feed rate but this increase is very small as compared to the increase in case of depth of cut and cutting speed.

Fig. 4
figure 4

Main effects plot of tool wear. a Cemented carbide. b TiN. c TiAlN. d PCD

Full size image

For better intuition and comparison purposes, the mean values of tool wear obtained from the main effect plots are given in Table 8. It is noticed that the mean values of tool wear for cemented carbide, TiN, TiAlN, and PCD coated tools were 57, 70, 45, and 106 μm, respectively, which clearly shows that TiAlN coated inserts exhibit least tool wear. TiAlN coated inserts exhibit on average ≈ 21% less tool wear than the uncoated cemented carbide inserts, ≈ 36% less tool wear than the TiN coated inserts, and ≈ 58% less tool wear than the PCD coated inserts. The better performance of TiAlN coated tools in terms of tool wear is expected because it is the most commonly used coated cutting tool in the industry and research for machining [51].

Table 8 Mean values of tool wear and workpiece surface roughness for all the four cutting inserts
Full size table

Figure 5 illustrates the main effect plots of workpiece surface roughness for all the four cutting inserts. The depth of cut has a linear positive effect on the workpiece surface roughness. After increasing the depth of cut from 1.0 mm, the abrupt increase in the workpiece surface roughness has been observed for all the cutting inserts. It is primarily because, at higher depth of cut, the workpiece-tool engagement is greater which not only causes to increase in the tool wear but workpiece surface roughness as well. Cutting speed and feed rate has a negative effect on workpiece surface roughness for all the cutting inserts. However, the decreasing effect is more prominent and significant for cutting speed. Cutting speed has shown a constant decrease in the workpiece surface roughness with an increase in the cutting speed which is according to the findings of Bouacha et al. [50].

Fig. 5
figure 5

Main effects plot of workpiece surface roughness. a Cemented carbide. b TiN. c TiAlN. d PCD

Full size image

Table 8 demonstrates the mean values of workpiece surface roughness for cemented carbide, TiN, TiAlN, and PCD coated inserts, which are 0.98, 0.81, 1.29, and 1.40 μm, respectively. TiN coated inserts exhibit the best workpiece surface roughness for the dry machining of the Al alloy 2024. This is due to the exceptional self-lubricating properties of the TiN coating [52,53,54] which causes to decrease in the workpiece surface roughness during the machining. TiN coated inserts exhibit on average ≈ 17% less workpiece surface roughness than the uncoated cemented carbide inserts, ≈ 37% less workpiece surface roughness than the TiAlN coated inserts, and ≈ 42% less surface roughness than the PCD coated inserts.

Comparing Figs. 4 and 5 at higher cutting speed and feed rate, the surface finish of the workpiece improves. However, this improvement in the surface finish is achieved at the cost of tool wear; the tool wear increased for all cutting inserts at higher values of cutting speed and feed rate. Therefore, higher machining time efficiency can be achieved at the cost of tool life for the dry machining of Al alloy 2024.

Table 6 shows that all the machining parameters (cutting speed, feed rate, and depth of cut) are highly significant (p < 0.05), but Pareto charts are also developed for better intuitive understanding. Figures 6 and 7 shows the Pareto charts of tool wear and surface roughness, respectively, for all the inserts. Depth of cut (DOC), cutting speed (CS), and feed rate (FR) is denoted by A, B, and C, respectively. Single terms like A, B, C show the linear significance on tool wear and workpiece surface roughness responses, while double terms like AA, BB, CC represent the nonlinear or quadratic significance on tool wear and workpiece surface roughness responses. However, the double terms like BC, AB, and AC demonstrate the two-way interaction significance on the desired responses.

Fig. 6
figure 6

Pareto charts of the tool wear. a Cemented carbide. b TiN. c TiAlN. d PCD

Full size image
Fig. 7
figure 7

Pareto charts of the surface roughness. a Cemented carbide. b TiN. c TiAlN. d PCD

Full size image

Figure 6 shows that the order of significance among the control variables for the tool wear for all the four types of inserts are the linear terms of depth of cut, cutting speed, and feed rate. Based on a reference line of significance at α = 5%, mostly the linear terms in the response surface regression model are significant to the response (tool wear) for all the inserts. But for the cemented carbide and PCD, a couple of nonlinear terms also have minor impact on the tool wear. Similarly, Fig. 7 represents that the single linear terms have the main impact in deciding the workpiece surface roughness, and the order of significance for different cutting inserts is different. However, based on the reference line of significance, certain nonlinear terms, and the interactions of control variables have a considerable impact on workpiece surface roughness.

3.4 Tool wear analysis

Each experiment was conducted for 30 min on the lathe machine, and then the tool wear was measured from CMM for all the 108 experiments (27 for each cutting tool type). As from the previous discussion, it is substantiated that the depth of cut is the most important parameter for tool wear, and feed rate is the least influential on tool wear. We chose experiment no. 3 from Table 4 (Fig. 8) and experiment no. 19 from Table 4 (Fig. 9).

Fig. 8
figure 8

CMM image for wear analysis of experiment no. 3

Full size image
Fig. 9
figure 9

CMM image for wear analysis of experiment no. 19

Full size image

Figure 8 shows the wear of all the cutting tools for experiment no. 3. It is seen that only the flank wear was observed in all the tools which are the most common wear type in machining tools [55]. This is typically due to the thermal softening of the cutting tool at high temperature during machining [56, 57]. Furthermore, it is quite clear from Fig. 8 that PCD has greater wear and TiAlN has the minimum. This trend is supported by the quantitative data in Table 4. It is imperative to note that in the case of PCD coated tools, the trend of wear mechanism is very strange as compared to the other tools; PCD showed a major adhesive failure (chip off) of the coating.

Similarly, Fig. 9 shows the wear of all the cutting tools for experiment no. 19. Flank wear is observed, here, also in all the cutting tools. The trend is also the same here; PCD showed the maximum wear while TiAlN showed the minimum. However, the magnitude of the wear is less for experiment no. 19 than the experiment no. 3. This is because experiment no. 3 was performed at a maximum depth of cut (which is most influential for tool wear in this study) while experiment no. 19 was conducted at maximum feed rate (which is least influential on tool wear in this study).

3.5 Response optimization

The selection of suitable machining parameters is a vital concern for the manufacturing industry to optimize the tool wear and the workpiece surface roughness. The response surface optimization method in Minitab® is employed to find suitable machining parameters. [56]. It is an extensively reported technique in estimating the best machining parameters against the output responses, and it gives us lower, target, and upper values of the output responses. The response optimizer function of the Minitab® gives us a composite desirability value on a scale of 0–1. The value which is closer to 1 is the more desirable and opposite for the value closer to zero [44, 58]. In this study, the goal was to achieve minimum tool wear and minimum workpiece surface roughness. The results for all the cutting inserts are given in Table 9. The minimum possible surface roughness is 0.45 μm for TiN coated cutting inserts at DOC of 0.5 (mm), CS of 80 (m/min), and FR of 0.123 (rev/min). The minimum possible tool wear is 14 μm for TiAlN coated cutting inserts at DOC of 0.5 (mm), CS of 92.52 (m/min), and FR of 0.110 (rev/min). Moreover, the high composite desirability values exhibit good confidence in the optimization results.

Table 9 Response optimization for tool wear and surface roughness of all inserts
Full size table

3.6 Effect of mechanical properties and adhesion on tool wear and workpiece surface roughness

Hardness and elastic modulus of the cutting inserts were evaluated by the nanoindentation method. Applied load and resulting displacements are continuously measured during the loading and unloading. Then, this data was utilized to plot load-displacement curves. Hardness and elastic modulus are extracted from these curves by Oliver and Pharr’s method [31, 59]. Load-displacement curves of all the cutting inserts are given in Fig. 10. Hardness and elastic modulus values of cemented carbide, TiN coated, TiAlN coated, and PCD coated cutting inserts are given in Table 10. PCD coated inserts have the highest hardness (≈ 35GPa) and elastic modulus (≈ 555GPa) as compared to the cemented carbide, TiN, and TiAlN coated inserts.

Fig. 10
figure 10

Load-displacement curves of cutting inserts. a Cemented carbide. b TiN coated. c TiAlN coated. d PCD coated

Full size image
Table 10 Results of adhesion and mechanical properties of the cutting inserts
Full size table

Adhesion of the TiN, TiAlN, and PCD coated tools was obtained from scratch tests. In the scratch test, when the normal load is applied to the specimen then micro-cracks are produced at the point of application along with the residual stresses. By increasing the applied load, the cracks are propagated along with the direction of the scratch. Once the residual stresses surpass the adhesion strength of the coating, then the coating failure occurs. Critical load at which coating completely scratch off is called the adhesion of the coating [60, 61]. Scratch images of TiN, TiAlN, and PCD coated insert is given in Figs. 11, 12, and 13, respectively, and their adhesion values are given in Table 10.

Fig. 11
figure 11

a Optical micrographs of the scratch test of the TiN coated cutting insert combined with the (b) graph of acoustic emission, critical load, and penetration depth

Full size image
Fig. 12
figure 12

(a) Optical micrographs of the scratch test of the TiAlN coated cutting insert combined with the (b) graph of acoustic emission, critical load, and penetration depth

Full size image
Fig. 13
figure 13

(a) Optical micrographs of the scratch test of the PCD coated cutting insert combined with the (b) graph of acoustic emission, critical load, and penetration depth

Full size image

Figure 11 shows the scratch images of TiN coated cutting insert along with critical load, scratch length, acoustic emission, and penetration depth. The extent of adhesive failure is recorded at four critical loads. At the first critical load (LC1), the coating has just started to scratch off. At the second critical load (LC2), the scratch has progressed, but the complete failure of TiN coating has not been initiated. At the third critical load (LC3), semi-circular cracks are evident along the whole scratch track, and in the final critical load (LC4), complete delamination of the coating occurs with complete exposure of the substrate. Such type of failure behavior of the TiN coating is according to the previous study [62]. Although the substrate is completely visible at LC4, the huge spikes of the acoustic emission occur at LC2 which indicates that the adhesion strength of the TiN coating should be recorded at this point. Furthermore, TiN coating observed a typical tensile cracking failure mode which has the characteristics of semi-circular crack propagation along the channel of the scratch [63, 64].

Figure 12 shows the scratch images of the TiAlN coated cutting insert along with critical load, scratch length, acoustic emission, and penetration depth graph. Like the TiN coating, TiAlN coating also showed the same failure events at different critical loads and failure modes. At the first critical load (LC1), the coating did not show any sign of chipping or delamination. At the second critical load (LC2), coating starts to chip off at several spots and completely delaminates at LC3 with semi-circular cracks indicating the tensile cracking failure mode [63, 64]. Since the substrate is completely visible at LC3 and corresponding acoustic emission peaks are also occurred at this point, thus, this point is recorded as the adhesion of the TiAlN coating.

Similarly, Fig. 13 shows the scratch images of the PCD coated cutting insert along with critical load, scratch length, acoustic emission, and penetration depth graph. At the first critical load (LC1), along the whole path of the scratch, the substrate becomes visible with clear spiral semi-circular cracks confirming the similar adhesion failure mode like the other two coatings. The strong acoustic signal is also observed at the LC1; thus, this point is recorded as the adhesion strength of the PCD. PCD coating fails at first critical load which indicates that it has very poor adhesion strength. In sum, from the scratch tests, TiAlN coated insert showed greater (≈ 15 N) coating adhesion as compared to the TiN coated insert (≈ 11 N) and PCD coated (≈ 7 N) insert.

It is a common convention that the tools which have high surface hardness would exhibit less tool wear (high wear resistance). H3/E2 ratio is also recently being reported to be a more suitable parameter in assessing the wear resistance of the materials [65,66,67]. Table 10 shows that H3/E2 ratio for PCD coated inserts is highest among cemented carbide, TiN, and TiAlN coated inserts. PCD coated inserts have approximately 85% more H3/E2 ratio than the TiAlN coated inserts and approximately 200% more H3/E2 ratio than the TiN coated inserts and cemented carbide inserts. Therefore, conventional wisdom dictates that the PCD coated inserts must have better wear resistance or less tool wear.

However, tool wear results (discussed in the previous sections) of uncoated cemented carbide, TiN, TiAlN, and PCD coated inserts indicate that the PCD coated inserts have the highest tool wear or poorest wear resistance. Those results suggest that the tool wear performance of the cutting inserts cannot be completely explained by hardness or H3/E2 ratio. Adhesion strength must also be considered in evaluating the wear resistance of the coatings.

Thus, to investigate the reason behind the highest tool wear or least wear resistance of PCD coated inserts despite having the higher mechanical properties, we measured the adhesion values of all the coated tools and concluded that TiAlN coated inserts showed the least tool wear owing to the highest adhesion strength, while PCD coated inserts displayed more tool wear because of the lowest adhesion strength. PCD coated inserts undergo extreme adhesive failure during the machining that big flakes of the coating chipped off from the tool, as can be observed in the tool wear images (Fig. 8). A detailed EDX analysis of the chip was conducted to verify these findings.

Workpiece surface roughness is also dependent on both the mechanical properties and adhesion strength of the coated tools. Cutting inserts which have high H3/E2 ratio and hardness show higher workpiece surface roughness. Poor adhesion will also play a role in surface quality. For example, PCD has greater hardness and very poor adhesion which results in the worst workpiece surface roughness in our study. Because when the wear debris of these cutting inserts comes in contact between the workpiece and tool, then it acts as a third body and causes for poor workpiece surface finish [68, 69]. Thus, workpiece surface roughness is decreased as the mechanical properties of the tools increase and the situation of the surface roughness becomes worse when that tool also has poor adhesion. However, there is an exception observed in our research regarding TiN coating, which is harder than cemented carbide, but it produces very good results of workpiece surface roughness because of its exceptional self-lubricating properties [52,53,54].

3.7 Chemical composition of the chips by EDX analysis

The optimum cutting parameters were used to collect the chips for EDX analysis. Workpieces were machined for a single stroke with all the cutting inserts (cemented carbide, TiN, TiAlN, and PCD coated), and then the chips were collected separately for EDX analysis. This analysis is important to evaluate the tool performance based on their mechanical properties (hardness) or adhesion strength. The findings of the EDX analysis helped to substantiate that the adhesion strength is also a major factor in assessing the machining performance.

The EDX analysis of the chips machined from the cemented carbide, TiN, TiAlN, and PCD coated inserts is demonstrated in Figs. 14, 15, 16, and 17, respectively. The detailed elemental composition is presented in Table 11. From Figs. 14, 15, and 16, it is seen that the peak of Al is higher in the spectrum than all the other elements. Al is the constituent element of the workpiece and thus has a higher % atomic composition. There is no evidence of tungsten (in case of cemented carbide insert) on the surface of the chip suggesting small wear of the uncoated cemented carbide inserts. Similarly, the absence of constituent elements (Ti and N) of the TiN and TiAlN coated inserts validate no abrupt adhesive failure of the coatings in their respective EDX spectrum and Table 11.

Fig. 14
figure 14

EDX analysis of the chips obtained from machining with cemented carbide cutting insert

Full size image
Fig. 15
figure 15

EDX analysis of the chips obtained from machining with TiN coated cutting insert

Full size image
Fig. 16
figure 16

EDX analysis of the chips obtained from machining with TiAlN coated cutting insert

Full size image
Fig. 17
figure 17

EDX analysis of the chips obtained from machining with PCD coated cutting insert

Full size image
Table 11 Elemental composition of the chips obtained from EDX analysis
Full size table

However, the EDX analysis of the chips machined from PCD coated insert showed strange behavior illustrated in Fig. 17. Significant visible black depositions of carbon can be observed in the scanning electron microscope (SEM) image and the strong carbon peak in the EDX spectrum validates this. This shows that intensive failure of the PCD coating occurs due to weak adhesion strength. None of the other coated inserts show such results and this led us to the conclusion that adhesion strength plays a substantial role in machining performance of the coated tools. PCD insert had very high hardness but it still performed poorly on both accounts of machining performance at optimum cutting parameters. The poor adhesion of the PCD insert causes the immediate adhesive failure of the coating when the tool comes in contact with the workpiece. Once failed, the ultra-hard wear debris of the PCD insert works as the damaging third body responsible for the poor surface finish of the workpiece [68, 69]. EDX analysis of the chips substantiated our experimental findings regarding the tool wear and the fact that adhesion strength of the coating and the mechanical properties of the surface coatings both are necessary for deciding the machining performance. Therefore, the research and industrial community in machining need to investigate these factors while deciding the tooling.

3.8 A comprehensive comparison of all the four cutting inserts

To comprehensively compare the machining performance of the cutting tools in dry machining of Al alloy 2024, all the properties and responses are summarized in Table 12. It consists of adhesion strength, hardness, elastic modulus, H3/E2 ratio, and the machining performance indicators, tool wear, and workpiece surface roughness. For a better intuition of the readers through a graphical display, a radar chart illustrates in Fig. 18, is created from the data given in Table 12. A radar chart is a powerful tool for comparing multivariant data in a two-dimensional graph when it is hard to interpret from tabular data [19].

Table 12 Summary of all cutting inserts’ properties and machining performance
Full size table
Fig. 18
figure 18

Radar chart showing the comprehensive comparison of cutting inserts

Full size image

Figure 18 compares the machining performance indicators for all the four cutting inserts based on the adhesion and mechanical properties. Despite having greater mechanical properties, PCD shows the worst machining performance results due to poor adhesion. TiAlN cutting insert performs superior in terms of tool wear, and TiN cutting insert displays excellent results of workpiece surface roughness. Thus, not a single tool can give desirable results. From the findings of this study, it is concluded that tool wear and workpiece surface roughness have inverse relations. Machining community must have to tradeoff between them. Reduction in tool wear will cost in terms of poor workpiece surface finish and vice versa.

4 Conclusions

Dry machining of aluminum alloy 2024 has been performed with four different cutting inserts (cemented carbide, TiN coated, TiAlN coated, and PCD coated) by varying three different machining parameters. DOE and RSM were conducted to optimize the machining parameters for tool wear and better workpiece surface roughness. Effect of mechanical properties and adhesion of the cutting inserts on tool wear and workpiece surface roughness is evaluated. The following conclusions have been drawn:

  1. 1.

    TiAlN coated cutting insert presented lowest tool wear than the uncoated cemented carbide, TiN, and PCD coated cutting inserts. TiAlN coated inserts exhibit on average ≈ 21%, ≈ 36%, and ≈ 58% less tool wear than the uncoated cemented carbide, TiN, and PCD coated inserts, respectively.

  2. 2.

    TiN coated inserts exhibit an overall minimum workpiece surface roughness than the TiAlN coated and the uncoated cemented carbide inserts. TiN coated inserts exhibit on average ≈ 17%, ≈ 37%, and ≈ 42% less workpiece surface roughness than the uncoated cemented carbide, TiAlN and PCD coated inserts, respectively.

  3. 3.

    Tool wear is less at low depth of cut, cutting speed, and feed rate while workpiece surface roughness is less at low depth of cut, higher cutting speed, and higher feed rate for aluminum alloy 2024.

  4. 4.

    Optimization results show that the minimum possible surface roughness is 0.45 μm for TiN coated cutting inserts at the DOC of 0.5 mm, CS of 80 m/min, and FR of 0.123 rev/min. The minimum possible tool wear is 14 μm for TiAlN coated cutting inserts at the DOC of 0.5 mm, CS of 92.52 m/min, and FR of 0.110 rev/min.

  5. 5.

    Nanoindentation and scratch test results show that the PCD coated inserts have higher hardness, elastic modulus, and while TiAlN coated inserts have higher adhesion as compared to the other cutting inserts.

  6. 6.

    Mechanical properties and adhesion of the coated tools are equally important in assessing the wear resistance and tool wear. Tool wear depends not only on the hardness or H3/E2 ratio but also on the adhesion. EDX analysis of the chips validated the experimental results of tool wear and thus cutting tools will have less wear if they have high mechanical properties and adhesion strength.

  7. 7.

    The highest hardness and H3/E2 ratio of the PCD inserts become its greatest disadvantage on both accounts of machining performance due to its poorest adhesion.

  8. 8.

    The self-lubricating nature of TiN wear debris is a prime reason for the best workpiece surface roughness it offers.

  9. 9.

    In machining of the Al 2024, the depth of cut should be kept minimum and cutting speed should be higher.

  10. 10.

    Research and industrial community in machining need to consider both the mechanical and adhesion properties in assessing and deciding the performance of the coated tools.

  11. 11.

    Extensive investigations are recommended for the research community to develop a comprehensive understanding of temperature and forces generated in the machining process. Further, the mechanical properties (hardness and H3/E2 ratio) and adhesion of the coatings need to be compared with the aforementioned forces in a range of operating temperatures.

  12. 12.

    Similarly, extensive research needs to be conducted in relating the failure mechanism of the coatings during machining with the microstructure of the coatings and structural and chemical properties of the interface between the coatings and tool substrate.