Sage Journals: Discover world-class research

Abstract

The present paper reports an experimental study on the fabrication and characterization of aluminum-oxide-reinforced functionally graded aluminum composites and optimization of drilling parameters on thrust force and average surface roughness using gray relational analysis. For this purpose, AA7075/Al₂O₃ functionally graded materials were produced with three layers that have different ratios of Al₂O₃ via high-temperature isostatic pressing and powder metallurgy method. Hardness behaviors of the layers were determined. Further, microstructural characterizations of the layers were carried out. Functionally graded materials were machined at dry cutting conditions with a 6 mm uncoated cemented carbide drill bit. The drilling experiments were carried out under different conditions such as point angles, helix angles, and feed rates, while the 25 m/min cutting speed was kept constant. The effects of factors on thrust force and surface roughness were evaluated using analysis of variance and gray relational analysis in full factorial experimental design. The minimum thrust force was measured at 221 N in a 140° point angle, 15° helix angle, and 0.075 mm/rev feed rate. Results of the gray relational analysis showed that feed rate was the dominant factor on thrust force and average surface roughness. Consequently, the highest and lowest gray relational grades were obtained at 0.899 and 0.374, respectively.

Keywords

Functionally graded material drilling thrust force surface roughness gray relational analysis

Introduction

Aluminum alloys (AAs) are preferred materials in engineering applications because of their high specific strength and stiffness as well as low density.^1,2 However, the main problem is that AAs have poor endurance in high temperatures. Among all AAs, the 7xxx series demonstrates the highest room-temperature strength. The 7xxx series can be strengthened by heat treatment through contained extremely fine and uniformly dispersed precipitates.^3,4 Adding a ceramic particle to AAs increases the mechanical properties at high temperature. Aluminum oxide (Al₂O₃), boron carbide (B₄C), and silicon carbide (SiC) ceramics are commonly used reinforcements in aluminum composites.^5–7 Al₂O₃ is one of the most important strengthening materials because of its excellent properties (e.g. high hardness, chemical stability, strength, corrosion resistant, low thermal expansion coefficient, cost, etc.).^8,9

Functional graded materials (FGMs) are composites in which the material composition gradually changes in a certain direction according to the desired properties.^10,11 Thanks to continuous change in their microstructures, FGMs are distinguished from conventional composites in high-temperature surface-wear resistance, increased adhesion at the ceramic–metal interface, and fracture toughness, along with reduced interfacial and thermal stress.¹² FGMs are used in many fields such as aerospace, defense, biomedical, nuclear, cutting tools, and electronics.^13,14 There are many different approaches to the production of FGMs such as chemical vapor deposition, centrifugal casting, spark plasma sintering, and powder metallurgy.¹⁵

Powder metallurgy (PM) is widely used in the production of FGMs due to its comprehensive control over the microstructure and ability to form shapes. In addition, shorter production times, simple processing equipment, low cost, and energy consumption are other advantages of PM.¹⁶

PM, which is a technique for producing parts near net shape, has four primary steps: (1) preparation of powder considering predesigned dispersion; (2) mixing of powder; (3) forming; (4) and sintering. However, secondary machining processes (e.g. drilling, turning, milling, etc.) are required when parts geometry has small holes, tiny walls, and fine finishing tolerance.¹⁷ Especially to reduce huge mold cost and to increase mold life, the geometrical elements (e.g. hole, screw, etc.) are formed secondary machining processes.¹⁸ Karabulut et al.¹⁹ studied the effect of reinforcement element type on surface roughness and chip formation to determine the drilling behavior of metal matrix composites. The authors fabricated three types of composites, which have 10 wt.% Al₂O₃, wt.% B₄C, wt.% SiC, via the powder metallurgy and hot-extrusion method. The authors further observed that better interfacial bonding of between reinforcement and matrix was Al₂O₃-aluminum samples. They also determined that better holes surface quality appeared in Al2O3-aluminum samples. The authors explained the reason as ductile machining with short chip formation.

Determining the machinability properties of materials is a complex and multifactorial problem. Multiresponse optimization methods are used to determine the optimum process parameters in several machining methods.²⁰ Gray relational analysis (GRA) is one of the effective multiresponse optimization methods used to evaluate the effects of factors and their levels.²¹ GRA creates optimized response input parameters, which have different weights of evaluation.²² Ranking of process parameters can also be accomplished by considering the effect of all response parameters.²³ In addition, the absence of complex formulas and calculations in GRA makes this technique easier and flexible. For these reasons, many researchers have used the GRA optimization method in their studies. Haq et al.²⁴ investigated the effect of machining parameters (i.e. cutting speed, feed rate, point angle) of Al/SiC metal matrix composites on cutting force, torque, and surface roughness using GRA and ANOVA. From this analysis, it was reported that point angle, cutting speed, and feed rate were effective in drilling with the contributions of 43.2%, 28.64%, and 26.21%, respectively. Babu et al.²⁵ examined the optimum drilling parameters of SiC-reinforced aluminum functionally graded composites using GRA. They fabricated experimental samples using the centrifugal casting method. They further reported that feed rate and drilling zone (as varying reinforcement rates) are the major factors with the contribution of 56.67% and 32.95% on the drilling process, respectively. Taşkesen and Kütükde²⁶ optimized the effect of the drilling parameters (i.e. feed rate, spindle speed, drill type) and reinforcement ratio on thrust force, torque, and surface roughness using GRA. The authors indicated that the most effective factor in machining was the reinforcement ratio. They also reported that surface roughness decreased with increasing the reinforcement ratio for the carbide tool. Rajeswari and Amirthagadeswaran²⁷ investigated the effect of spindle speed, feed rate, depth of cut, and ratio of reinforcements on the milling of Al7075-based SiC-Al₂O₃-reinforced composites. The authors used ANOVA to obtain the most influencing parameters on thrust force and torque. According to the ANOVA, the ratio of SiC and spindle speed were dominant factors. The authors observed that increasing the feed rate increased thrust force and torque. Tosun and Muratoglu²⁸ investigated the effect of drill types (i.e. HSS, TiN-coated HSS, carbide) and point angles (90°, 118°, 130°) on drilled surface damage during drilling of 17 wt.% SiC-reinforced AA2124 composite material. The results of this study showed that solid carbide drills produce better surface roughness compared with TiN-coated HSS and HSS drills; further, a 130° point angle was suggested due to reduced damage zones with an increase of point angles of drill bits.

Cutting tool wear negatively affects machined surface quality, hole diameter tolerance, and geometry. Analysis of tool wear is important for better machining operation. In particle-reinforced metal-based composites, especially, aluminum alloys have a greater tendency to stick to the cutting tool under dry machining conditions.²⁹ During the cutting process, a built-up edge (BUE) is formed as a result of the soft workpiece material sticking to the cutting edge of the cutting tool.³⁰ Several factors such as type, percentage and size of reinforcement, and tool can affect tool wear when drilling metal matrix composites geometry (e.g. point angle, helix angle, etc.), and drilling parameters (e.g, feed rate, cutting speed, etc.).^31–34 Choosing the right parameters is crucial for improving the process efficiency of metal matrix composites.³⁵

The literature includes many experimental studies on drilling properties of metal matrix composites in terms of the evaluation of surface roughness, thrust force, and tool wear. However, comprehensive studies on the drilling properties of FGMs, to our best knowledge, were not observed. Determining the drilling parameters and drill geometry of FGMs is of great importance for processing cost and material quality since the ratio of reinforcement in it varies according to the different zones.

The main aim of this research is to ascertain the mechanical and drilling performance of AA7075/Al₂O₃ FGMs. For this purpose, three layered AA7075/(20-40-60 wt.% Al₂O₃) FGMs were produced by high-temperature isostatic pressing and powder metallurgy method. Hardness properties of the layers were tested by the Vickers hardness-measurement technique, and microstructural investigations were carried out to determine metallurgical properties using SEM. Drilling experiments were done to examine the impact of the parameters (i.e. point angle, helix angle, and feed rate) on thrust force (F_z), average surface roughness (R_a), and tool wear. The experimental results were reported using ANOVA. The GRA method was used to evaluate the effects of tool geometry and feed rate on the thrust force and surface roughness parameters together.

Material and methods

Fabrication of FGMs

FGMs were fabricated using high-temperature isostatic pressing and powder metallurgy methods. AA7075 was used as a matrix material, and Al₂O₃ was used as a reinforcement material. The average particle sizes of AA7075 and Al₂O₃ were 150 and 10 µm, respectively. In sample production, first, AA7075 and Al₂O₃ powder materials were weighed on a weighing instrument according to their weight percentage. Then, the prepared AA7075/Al₂O₃ powders were mixed in a Turbula mixer for 45 min to prevent agglomeration. The mixed powders were layered in a lubricated die starting with AA7075/Al₂O₃ 20 wt.% at the bottom and finishing AA7075/Al₂O₃ 60 wt.% at the top. The hot-pressing die was fabricated hot work steel (AISI 2344) and hardened with a heat treatment process. The AA7075/Al₂O₃ powders were pressed at 10 MPa for 2 min at room temperature. After pre-pressing, the compact FGMs were heated at 450°C for 30 min. Then, the FGMs were pressed at the condition of 325 MPa and 450°C for 30 min. They were cooled at room temperature. Five samples were produced in this study. Three of them were used for mechanical and microstructural confirmation, and two were used for drilling experiments. The AA7075/Al₂O₃ FGMs preparation is summarized in Figure 1. FGMs were produced in 60 × 60 × 12 mm dimensions with each layer at 4 mm.

Figure 1.

Schematic diagram of production of functionally graded materials.

Mechanical and microstructural analysis

The microstructure and distribution of Al₂O₃ in the metal matrix was observed using a scanning electron microscope (SEM). For this purpose, first, three test specimens were sectioned and then prepared according to standard metallographic sample preparation techniques. The microhardness of FGMs was measured with a Vickers hardness (HV) test machine under 1 kg load applied for 15 s. Microhardness measurements were performed at five different regions for each layer and repeated on three different samples. The average hardness values of the layers were calculated by taking the average of the results.

Experimental setup

Drilling trials were done by a CNC vertical machining center (Johnford VMC-550 Fanuc Series O-M). A “Kistler 9272” dynamometer with “Dynoware” software was used to measure the drilling forces (Fx, Fz, and Fz). A Schunk BT40 holder was used for the drilling. The drills were specially manufactured using cementite carbide bar material (Ceratizit; P-line, submicrometer carbide grade code: CTS12D) by KARCAN Co. cutting tools with a diameter of 6 mm to examine the impact of drilling parameters (i.e. point angle, helix angle, feed rate) on thrust force, average surface roughness, and tool wear. Drilling experiments were carried out from top layers of FGMs (60 wt.% Al₂O₃) to bottom layers (20 wt.% Al₂O₃) in dry conditions, which are shown in Figure 2. The average surface roughness of a drilled hole was measured using the Mitutoyo (SJ-410 Series) surface roughness test machine according to the ISO 1997 test standard. The surface roughness of holes was measured in the machining direction. Each hole was measured four times by rotating the hole position 90°. Tool wear of drill bits was observed using a Nikon ShuttlePix microscope. Tool wear images were taken at 25× and 225× magnification. Unit measurement lengths are given on the images.

Figure 2.

Experimental setup of the drilling process.

Experimental design for drilling

The experimental drilling tests were made by using a full factorial (FF) design. FF enables testing of all possible combinations of the levels of each factor. FF allows us to examine the influence of each factor on the response variable. Also, it is one of the effective methods used to determine the impacts of interactions between factors on the response variable. In this study, the control parameters such as point angle, helix angle, and feed rate were varied to investigate the effect of tool geometry on thrust force (F_z) and surface roughness (R_a). Three different levels were selected to determine the effect of control parameters, as shown in Table 1.

Table 1.

Control parameters and their levels.

Control parameters	Level
Control parameters	1	2	3
Point angle (ε),°	120	130	140
Helix angle (ω),°	0	15	30
Feed rate (f_n), mm/rev	0.075	0.1	0.125

Drilling is a multifunctional complex operation. In order to clearly determine the effect of tool geometry on the drilling of functionally graded composites, tool diameter (6 mm) and cutting speed (25 m/min) were taken as constants. Drilling experiments were done in dry conditions. A total of 27 ([level number]^{factor number} = 3³) experiments were conducted within the scope of full factorial experiment design.

GRA

GRA is a multiresponse optimization technique. The GRA determines the basic relationships between the factors of a system that has less information on data and cannot be modeled due to uncertainties.³⁶ There is no complex formulas and calculations in the field of gray relational analysis that make this technique easy and feasible.²⁴ GRA is also a grading, classification, and decision-making technique.

The calculation stages of the gray relational analysis method are as follows:

Stage 1: Determining the reference and the factor series

The reference series of $x_{0} (n)$ formed by k variables and the factor series $x_{i} (n)$ formed by k variables to be compared with each other are determined.

\begin{matrix} x_{0} (n) = (x_{0} (1), x_{0} (2), x_{0} (3), \dots, x_{0} (k)) \\ n = 1, 2, 3, \dots, k \end{matrix}

(1)

\begin{matrix} x_{i} (n) = (x_{i} (1), x_{i} (2), x_{i} (3), \dots, x_{i} (k)) \\ n = 1, 2, 3, \dots, k i = 1, 2, 3, \dots, m \end{matrix}

(2)

$k, m \in N$

$x_{0} (n)$ : The reference series.

$x_{i} (n)$ : The factor series.

Stage 2: Normalization of the experimental data.

Experimental data are normalized in series to make comparisons between series, which have different measurement units. Normalization is between $1 \leq x_{i} \leq 0$ .

The larger, the better

x_{i} (n) = \frac{x_{i}^{0} (n) - min x_{i}^{0} (n)}{max x_{i}^{0} (n) - min x_{i}^{0} (n)}

(3)

The smaller, the better

x_{i} (n) = \frac{max x_{i}^{0} (n) - x_{i}^{0} (n)}{max x_{i}^{0} (n) - min x_{i}^{0} (n)}

(4)

The nominal, the better

x_{i} (n) = \frac{| x_{i}^{0} (n) - x^{0} |}{max x_{i}^{0} (n) - x^{0}}

(5)

Stage 3: Gray relational coefficient.

The gray relational coefficient (ε) is used to describe the relationship between ideal and normalized results. $x_{0} (n)$ is the reference series; $x_{i} (n)$ is the comparison series. In the equation (9) below, ε (n) is the gray relational coefficient at the nth experiment. The identification coefficient (ζ) is determined in the range $0 \leq ζ \leq 1$ and is usually taken as 0.5.³⁷

Δ_{0 i} (n) = | x_{0} (n) - x_{i} (n) |

(6)

Δ_{\min} = mi n_{i} mi n_{k} | x_{0} (n) - x_{i} (n) |

(7)

Δ_{\max} = ma x_{i} ma x_{k} | x_{0} (n) - x_{i} (n) |

(8)

ε (n) = \frac{Δ_{\min} + ζ Δ_{\max}}{Δ_{0 i} (n) + ζ Δ_{\max}}

(9)

Stage 4: Calculation of the gray relational grade (GRG).

The GRG is a measure of the geometric similarity between the series $γ (x_{0}, x_{i})$ and the reference series in a gray system; it indicates the relationship between the series. The gray relational grade demonstrates how similar the comparison series is to the reference series. It means that the greater gray relational degree value, the stronger the relationship between $x_{i}$ and $x_{0}$ . The gray relational degree is calculated by multiplying the gray relational coefficient with the weight value $(W_{i} (n))$ of the criteria importance.³⁸

γ (x_{0}, x_{i}) \frac{1}{k} \sum_{n = 1}^{k} ε (x_{0} (n), x_{i} (n) . W_{i} (n))

(10)

Stage 5: Ranking of the gray relational grade.

It takes the first place with the maximum gray relational degree and ranks from the value with the higher gray relational coefficient to the lower one.

Stage 6: Determination of the optimal parameters with ANOVA.

ANOVA is used to determine the factor that has a significant effect on the gray relational degree.

Stage 7: Prediction of the optimum factor levels.

The total mean values are calculated based on GRG response values to the determine optimum parameter levels. The factor with the highest total mean value among the factors gives the optimum level.³⁹

Results and discussions

Mechanical and microstructural properties

AA7075-based FGMs reinforced with Al₂O₃ (20-40-60 wt.%) were produced by high-temperature isostatic pressing and powder metallurgy method. The microstructure of the powders, which are used in the production of FGMs, is presented in Figure 3.

Figure 3.

SEM image and EDX of powders used for fabrication of FGMs: (a) AA7075 and (b) Al₂O₃.

It can be seen that aluminum powders have a spherical shape, while Al₂O₃ ceramics powders have an angular shape. The SEM image of the microstructure of the produced FGM sample is given in Figure 4. Formation of a metallurgical bond between the layers allowed for a continuous transition from one layer to another.⁴⁰ The increase of Al₂O₃ caused an increase in the tendency of the ceramic particles to agglomerate. Due to the agglomeration, a smooth interface bond between the ceramic–metal matrix could not be established. Particle distribution was better in the layer containing Al₂O₃ 20 wt.%. However, full densification in the sample was not observed. The porosity was increased with an increasing amount of reinforcement.

Figure 4.

Microstructure of AA7075-Al₂O₃ FGMs: (a) between top and middle layer, (b) top layer of FGM, (c) between middle and bottom layer, (d) middle layer of FGM, and (e) bottom of FGM.

The microhardness of the FGM layers was evaluated. The influence of the amount of reinforcement ceramics on the hardness of layers is shown in Figure 5. The average microhardness of the layers containing 20 wt.% Al₂O₃, 40 wt.% Al₂O₃, and 60 wt.% Al₂O₃ was measured at 105.8, 126.4, and 102.8 HV, respectively. The hardness of all layers was significantly higher than AA7075 (78 HV)²⁶ hardness. The increasing microhardness of these layers with the Al₂O₃ content increasing is related to the dispersion strengthening.⁴¹ However, it was observed that the hardness value of the layer with 40 wt.% Al₂O₃ constituted a turning point. The decrease in the hardness of 60 wt.% Al₂O₃ layer is due to the agglomeration of reinforcing particles and higher porosity. In previous studies, this reduction was attributed to the poor bonding of the matrix and reinforcement interface.⁴¹

Figure 5.

Hardness of the layers.

Thrust force optimization

Thrust force is a primary indicator of machinability on drilling materials. Usually, the high drilling forces, which occur in the machining process, cause delamination between layers and failure input and output of composite holes. Therefore, thrust force must be reduced as far as possible. Choosing suitable tool geometry (e.g. point angle, helix angle, etc.) and cutting parameters (e.g. feed rate, cutting speed, etc.) can be effective in reducing thrust force.⁴²

In this study, the drilling force components in Figure 6 (F_x, F_y, and F_z) were measured for all combinations of factors and their levels. F_x and F_y occur in radial directions. The radial forces in the two cutting edge balance each other. Therefore, F_z is considered as the drilling force in this study. F_z is called the thrust force applied in the Z direction.⁴³

Figure 6.

Drilling force components.

Values of the F_z measured during the machinability experiments of FGMs are presented in Table 4. Parts of the measurement data graphic are shown in Figure 7, that is, that the thrust force gradually increases during the transition from one layer to another. When changes in thrust forces were examined, it was determined that thrust forces increased ∼25% on the first interface transition and ∼27.4% on the second interface transition compared with input mean force. Three main factors are believed to contribute to this increase. First, the fact that the cutting tool is confronted with more density surface as it moves from one layer to the next, and with the increasing amount of chip volume caused increased cutting forces. Second is the effect of increased friction depending on the increased hole depth. Third is the change of tool geometry due to tool wear and BUE.

Figure 7.

Parts of the Dynoware software data graphic by positions of the drill bit.

The mean value of the F_z was used to evaluate the results. ANOVA analysis was carried out using JMP7 software to obtain the influence of variables on the drilling characteristics. The results of ANOVA for the thrust forces are shown in Table 2. The investigation of ANOVA was used for the significance level of p = 0.05, that is, a 95% assurance level. If p is equal to 0.05 or less, it is possible to deduce that the influence of the factor is statistically significant.⁴⁴ The results of the ANOVA for thrust forces supported a strong linear relationship in the drilling models. R-square (R²) value of the analysis was 0.9177. R² is the variance ratio of the dependent variable that can be predicted from the independent variable.⁴⁴ Based on the ANOVA results, the most influential parameter on the thrust force was feed rate with a contribution of 27.05%.

Table 2.

ANOVA for thrust force.

Source	DF	Sum of squares	Mean square	F ratio	p-Value	Contribution rate (%)
Point angle (ε)	2	58,688.9	29,344.43	11.518	0.004	23.69
Helix angle (ω)	2	27,929.3	13,964.66	5.481	0.032	11.27
Feed rate (f_n)	2	67,004.7	33,502.37	13.150	0.003	27.05
Point angle × helix angle	4	46,727.3	11,681.84	4.585	0.032	18.86
Point angle × feed rate	4	16,880.4	4220.09	1.656	0.252	6.81
Helix angle × feed rate	4	10,109.8	2527.45	0.992	0.464	4.08
Error	8	20,382.0	2547.7			8.23
C. Total	26	247,722.5				100

Also, point angle and helix angle were effective with contributions of 23.69% and 11.27%, respectively. Small changes in the factors with a high percent contribution will have a major impact on hole drilling. The means of F_z for each level of drilling parameters are shown in Figure 8. It is observed that an increase in feed rates increases the F_z in the drilling of FGMs, which was demonstrated in previous studies.^45,46 In parallel with previous studies, it was investigated that the F_z decreases with the increase of the point angle.²⁵ Helix angles of 0° and 30° have a similar effect on thrust forces. The lowest thrust force was obtained at a 15° helix angle.

Figure 8.

Thrust force means of factors.

Surface roughness optimizations

Surface roughness is an important criteria for evaluating a material’s machinability. Machined surface quality affects material properties such as abrasion, fatigue strength, creep, and corrosion resistance.¹⁹ The results of ANOVA for the surface roughness are shown in Table 3. As a result of ANOVA, R² was calculated as 0.9517. The analysis indicates that the feed rate was the highest influential parameter on surface roughness with a 41.09% contribution rate. Also, point angle and helix angle were effective with a contribution of 16.40% and 6.03%, respectively. The interaction of point angle with the other factors were effective on the surface roughness. However, the interactions of the helix angle and feed rate did not demonstrate a statistical effect on surface roughness.

Table 3.

ANOVA for surface roughness.

Source	DF	Sum ofsquares	Mean square	F ratio	p-Value	Contributionrate (%)
Point angle (ε)	2	0.01841	0.00921	13.6098	0.003	16.40
Helix angle (ω)	2	0.00677	0.00338	5.0022	0.039	6.03
Feed rate (f_n)	2	0.04612	0.02306	34.0904	0.000	41.09
Point angle × helix angle	4	0.01582	0.00396	5.8469	0.017	14.09
Point angle × feed rate	4	0.01279	0.00320	4.7269	0.030	11.39
Helix angle × feed rate	4	0.00693	0.00173	2.5624	0.120	6.18
Error	8	0.00541	0.00068			4.82
C. Total	26	0.11227				100

In parallel with the previous study,³² it was observed that R_a increased with an increase of feed rates and decreased with a decrease of point angles, as shown in Figure 9. Maximum R_a was observed at a 15° helix angle as 0.365 µm; further, it was determined that the R_a reduced at the values below or above this value.

Figure 9.

Surface roughness means of factors.

Mathematical modeling

In this study, mathematical models were created to estimate the value of thrust force and surface roughness according to control factors. The models developed can also be used to investigate the effects of significant factor parameters on the output response. Mathematical models created according to ANOVA analysis are given in equations (11) and (12). The mathematical models were verified by efficiency measures (R²) of 0.92–0.95 for thrust force and surface roughness, respectively. The comparative thrust force and surface roughness values obtained with the regression equations and experimentally are given in Figures 10 and 11, respectively. The graphs show that mathematical equations output and experimental outputs are close to each other with a trend in high harmony.

\begin{matrix} Thrust Force = 14 + (0.04 \times Point angle) \\ + (47.8 \times Helix angle) + (1590 \times Feed rate) \\ - (0.364 \times Point angle \times Helix angle) \end{matrix}

(11)

\begin{matrix} Surface Roughness = 0.116 - (0.00013 \times Point angle) \\ + (0.0009 \times Helix angle) + (4.65 \times Feed rate) \\ + (0.000008 \times Point angle \times Helix angle) \\ - (0.0258 \times Point angle \times Feed rate) \end{matrix}

(12)

Figure 10.

Experimental and regressional thrust force results.

Figure 11.

Experimental and regressional surface roughness results.

Multiresponse optimization with GRA

In this section, gray relational analysis was used to determine the optimum thrust force and surface roughness. Gray relational analysis data are given in Table 4. “The smaller-the better” method was used to normalize the result parameters (equation (4)) because the F_z and R_a values are desired lower. Deviation orders $(Δ_{0 i} (n))$ were calculated using equation (6). Gray relational coefficients were calculated according to equation (9), and the identification coefficient $(ζ)$ was taken at 0.5. The GRGs were computed using equation (10). Weight value $W_{i} (n)$ expresses the given importance of the response orders; for the gray relational grade, the F_z was taken 50%, and R_a was taken at 50%. In the GRA, the highest GRG was gained in the 22nd experimental array with the value of 0.899, and the lowest GRG was gained in the ninth experimental array with the value of 0.374.

Table 4.

Results of gray relational analysis.

	Factors			Experimental		Normalization		Gray relationcoefficient		GRG	Rank
Exp. no	Pointangle (ε)(°)	Helixangle (ω)(°)	Feed rate(f_n) (mm/rev)	Thrustforce F_z(N)	SurfaceroughnessR_a (µm)	Thrustforce F_z(N)	SurfaceroughnessR_a (µm)	ThrustforceF_z(N)	SurfaceroughnessR_a (µm)	GRG	Rank
1	120	0	0.075	263.78	0.235	0.895	0.561	0.826	0.533	0.679	9
2	120	0	0.100	333.97	0.180	0.722	0.800	0.642	0.715	0.678	10
3	120	0	0.150	469.43	0.341	0.388	0.101	0.450	0.357	0.404	26
4	120	15	0.075	259.33	0.315	0.906	0.215	0.841	0.389	0.615	18
5	120	15	0.100	306.08	0.254	0.790	0.481	0.705	0.491	0.598	19
6	120	15	0.150	407.24	0.339	0.541	0.111	0.522	0.360	0.441	25
7	120	30	0.075	405.99	0.166	0.544	0.863	0.523	0.785	0.654	14
8	120	30	0.100	473.62	0.218	0.378	0.636	0.446	0.579	0.512	21
9	120	30	0.150	627.00	0.296	0.000	0.296	0.333	0.415	0.374	27
10	130	0	0.075	267.59	0.195	0.885	0.736	0.813	0.655	0.734	7
11	130	0	0.100	338.01	0.134	0.712	1.000	0.634	1.000	0.817	3
12	130	0	0.150	443.21	0.304	0.453	0.264	0.477	0.404	0.441	24
13	130	15	0.075	221.97	0.248	0.998	0.506	0.995	0.503	0.749	5
14	130	15	0.100	249.65	0.284	0.929	0.350	0.876	0.435	0.655	13
15	130	15	0.150	325.08	0.365	0.744	0.000	0.661	0.333	0.497	22
16	130	30	0.075	316.19	0.215	0.766	0.650	0.681	0.588	0.635	16
17	130	30	0.100	393.10	0.227	0.576	0.596	0.541	0.553	0.547	20
18	130	30	0.150	225.70	0.361	0.988	0.014	0.977	0.336	0.657	12
19	140	0	0.075	269.32	0.160	0.881	0.888	0.808	0.817	0.812	4
20	140	0	0.100	314.53	0.224	0.770	0.610	0.685	0.562	0.623	17
21	140	0	0.150	413.38	0.283	0.526	0.353	0.513	0.436	0.475	23
22	140	15	0.075	221.00	0.164	1.000	0.873	1.000	0.797	0.899	1
23	140	15	0.100	254.98	0.207	0.916	0.683	0.857	0.612	0.734	6
24	140	15	0.150	315.82	0.197	0.766	0.727	0.682	0.647	0.664	11
25	140	30	0.075	222.11	0.169	0.997	0.849	0.995	0.768	0.881	2
26	140	30	0.100	244.60	0.226	0.942	0.602	0.896	0.556	0.726	8
27	140	30	0.150	314.18	0.210	0.770	0.671	0.685	0.603	0.644	15

In addition, ANOVA was used to obtain the effects of factors on GRA responses. R² was calculated as 0.9232. Table 5 shows the feed rate, point angle, and helix angle, which influenced the GRG values with percentages of 47.49%, 24.73%, and 0.63%, respectively. These results show that the feed rate was the most effective parameter on the GRA value.

Table 5.

ANOVA for GRA.

Source	DF	Sum of squares	Mean square	F ratio	p-Value	Contribution rate (%)
Point angle (ε)	2	0.12554867	0.0627743	12.8924	0.0031	24.73
Helix angle (ω)	2	0.00318867	0.0015943	0.3274	0.73	0.63
Feed rate (f_n)	2	0.24108867	0.1205443	24.7571	0.0004	47.49
Point angle × helix angle	4	0.03869867	0.0096747	1.987	0.1896	7.62
Point angle × feed rate	4	0.01974133	0.0049353	1.0136	0.455	3.89
Helix angle × feed rate	4	0.04046133	0.0101153	2.0775	0.1759	7.97
Error	8	0.03895267	0.004869			7.67
C. Total	26	0.50768				100

GRA response values were determined. The parameter levels are shown in Table 6. The value that has the maximum response value among the levels gives the optimal parameters for the experimental design. The optimum response was attained in the array of “ε3-ω2-f_n1.” The values were 0.71756, 0.65022, and 0.73978, respectively. Total mean value of the GRA was calculated as 0.63500.

Table 6.

Response table for GRA.

Source	Level 1	Level 2	Level 3	Delta	Rank
Point angle (ε)	0.55056	0.63689	0.71756	0.16700	2
Helix angle (ω)	0.62922	0.65022	0.62556	0.02467	3
Feed rate (f_n)	0.73978	0.65444	0.51078	0.22900	1

Total mean value of the GRA = 0.63500

Tool wear

Tool wear is an expected situation in machining and drilling processes. In this section, the effects of tool geometry (point angle and helix angle) on the tool wear in the drilling of AA7075/Al₂O₃ FGMs were evaluated. In order to understand the effects of tool geometry on tool wear, the feed rate was kept constant as f_n: 0.075 mm/rev. Figure 12 shows the tool wear images formed at different point angles (120°, 130°, 140°) at a constant 30° helix angle. The high adhesion tendency of the matrix within the structure and the high hardness of the reinforcement particles caused the formation of built-up edge (BUE) on the cutting tool rake face. Similar BUE formation was detected at all point angles. BUE tended to increase in the cutting-edge cutting line toward the center of the drill bit. Also, it was observed that the aluminum matrix was deposited on the chisel edge. When the buildup on the chisel edge length was measured, the maximum value was observed as 0.785 mm at the 120° point angle. This was attributed to narrow tool geometry in previous literature studies, which caused high friction between the tool and workpiece.^34,35 Figure 13 shows the tool-wear images formed at different helix angles (0°, 15°, 30°) at a constant 120° point angle. BUE and buildup on the chisel edge were observed at all helix angles. According to the measurements, it was determined that the buildup on chisel edge lengths were 0.425, 0.310, and 0.778 mm, respectively. The lowest buildup on the chisel edge was observed at a 15° helix angle. It is known that tool wear is related to cutting force. Previous studies proved that the increase in cutting force increases the machining temperature, which increases the tendency of the workpiece material to stick to the tool.^34,47,48

Figure 12.

Images of tool wear according to point angle: (a) 120°, (b) 130°, and (c) 140°.

Figure 13.

Images of tool wear according to helix angle: (a) 0°, (b) 15°, and (c) 30°.

Conclusions

In this study, the effect of tool geometry and drilling parameters on Al₂O₃-reinforced FGMs was investigated. To our best knowledge, there is not enough study in the literature about the drilling processes of materials with varying reinforcement ratios throughout the drill holes. The investigation of the drilling process of functionally graded materials is important for the literature. Al₂O₃-reinforced functionally graded aluminum composites were fabricated by powder metallurgy. The microstructure of composites was investigated by SEM images. Thrust force, surface roughness, and tool wear were taken into consideration as performance criteria. Full factorial design was used in experimental design, and the results were analyzed by using ANOVA and GRA. The results obtained in the study are summarized below.

The SEM image of the FGMs indicate that Al₂O₃ powders were not uniformly distributed in the layers of the composite. It was also observed that the porosity and agglomeration increase with an increasing amount of reinforcement.

The hardness was affected by the amount of Al₂O₃ in the layers. Maximum hardness was measured as 126.4 HV in the 40 wt.% Al₂O₃-reinforced layer.

Based on ANOVA, feed rate was the most effective parameter on thrust force with a contribution of 27.05%. In addition, the effects of the point angle and helix angle were 23.69% and 13.01%, respectively. F_z increased when feed rates increased. However, it decreased with an increasing of point angle. The minimum F_z was measured at 221 N in a 15° helix angle.

The ANOVA indicated that the feed rate was the most dominant factor on surface roughness with a contribution of 41.09%, followed by point angle 16.40% and helix angle 6.03%.

Regression equations developed by input parameters affecting thrust forces and surface roughness were obtained at high R² such as 0.82 and 0.95. Accuracy of the mathematical equation has been proved by graphs comparing experimental and mathematical data.

According to the GRA, the “ε3-ω2-f_n1” set has the maximum response value among the levels and gives the optimal parameters for the experimental design. The optimum results of GRA were found as R_a = 0.164 µm and F_z = 221.0. In addition, the significance of the factors on the GRG were the feed rate, point angle, and helix angle, respectively.

A built-up edge and built-up chisel edge were observed in all cutting tools. Optical images showed that the minimum built-up chisel edge length (0.310 mm) occurred at the 15° helix angle.

Determining the appropriate tool geometry and drilling parameters in the machining of FGMs prevents material waste and reduces production costs. Therefore, determining the drilling behavior of FGMs has great importance in terms of cost, time, and final product quality. This study fills a wide gap in the determination of the appropriate tool geometry and drilling parameter for drilling FGMs with an uncoated carbide drill in dry conditions. The effect of tool coating and cutting fluids on drilling behavior can be tested for further investigation.

Footnotes

Acknowledgements

Thanks to the OYP program for the financial support and KARCAN Co. for support in the research.

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 iDs

Tuğçe Yıldız

Gökhan Sur

References

Rouniyar

Shandilya

. Fabrication and experimental investigation of magnetic field assisted powder mixed electrical discharge machining on machining of aluminum 6061 alloy. Proc IMechE, Part B: J Engineering Manufacture 2019; 233: 2283–2291.

Hamilton

Kopyściański

Węglowska

, et al. Modeling, microstructure, and mechanical properties of dissimilar 2017A and 5083 aluminum alloys friction stir welds. Proc IMechE, Part B: J Engineering Manufacture 2019; 233: 553–564.

, et al. Influence of particle size and spatial distribution of B4C reinforcement on the microstructure and mechanical behavior of precipitation strengthened Al alloy matrix composites. Mater Sci Eng A 2016; 675: 421–430.

Yao

, et al. Surface integrity and fatigue behavior in shot-peening for high-speed milled 7055 aluminum alloy. Proc IMechE, Part B: J Engineering Manufacture 2017; 231: 243–256.

Samer

Andrieux

Gardiola

, et al. Microstructure and mechanical properties of an Al-TiC metal matrix composite obtained by reactive synthesis. Compos Part A Appl Sci Manuf 2015; 72: 50–57.

Karabulut

. Optimization of surface roughness and cutting force during AA7039/Al2O3 metal matrix composites milling using neural networks and Taguchi method. Measurement 2015; 66: 139–149.

Sasikumar

KSK

Arulshri

Ponappa

, et al. A study on kerf characteristics of hybrid aluminium 7075 metal matrix composites machined using abrasive water jet machining technology. Proc IMechE, Part B: J Engineering Manufacture 2018; 232: 690–704.

Sajjadi

Ezatpour

Beygi

. Microstructure and mechanical properties of Al-Al2O3 micro and nano composites fabricated by stir casting. Mater Sci Eng A 2011; 528: 8765–8771.

Yehia

Hakim

El-Assal

. Effect of the Al2O3 powder addition on the metal removal rate and the surface roughness of the electrochemical grinding machining. Proc IMechE, Part B: J Engineering Manufacture 2020; 234: 1538–1548.

10.

Gupta

Talha

. Recent development in modeling and analysis of functionally graded materials and structures. Prog Aerosp Sci 2015; 79: 1–14.

11.

Mahdavinejad

Khajeafzali

. Fabrication and experimental investigation of arc erosion behavior in W/Cu functionally graded composites. Proc IMechE, Part B: J Engineering Manufacture 235(3): 394–405.

12.

Choy

Sun

C-N

Leong

, et al. Compressive properties of functionally graded lattice structures manufactured by selective laser melting. Mater Des 2017; 131: 112–120.

13.

Naebe

Shirvanimoghaddam

. Functionally graded materials: a review of fabrication and properties. Appl Mater Today 2016; 5: 223–245.

14.

Esen

. Dynamic response of functional graded Timoshenko beams in a thermal environment subjected to an accelerating load. Eur J Mech A Solids 2019; 78: 103841.

15.

Kieback

Neubrand

Riedel

. Processing techniques for functionally graded materials. Mater Sci Eng A 2003; 362: 81–106.

16.

Mahamood

Akinlabi

. Functionally graded materials. Porto Alegre, Brazil: Springer, 2013.

17.

Cogun

Esen

Genc

, et al. Effect of powder metallurgy Cu-B4C electrodes on workpiece surface characteristics and machining performance of electric discharge machining. Proc IMechE, Part B: J Engineering Manufacture 2016; 230: 2190–2203.

18.

Ramulu

Rao

Kao

. Drilling of (Al2O3) p/6061 metal matrix composites. J Mater Process Technol 2002; 124: 244–254.

19.

Karabulut

Gökmen

Çinici

. Study on the mechanical and drilling properties of AA7039 composites reinforced with Al2O3/B4C/SiC particles. Compos Part B Eng 2016; 93: 43–55.

20.

Soorya Prakash

Gopal

Karthik

. Multi-objective optimization using Taguchi based grey relational analysis in turning of Rock dust reinforced Aluminum MMC. Measurement 2020; 157: 107664.

21.

Zheng

Liu

Guo

, et al. A new method for optimizing process parameters of active measurement grinding based on grey target decision making. Proc IMechE, Part C: J Mechanical Engineering Science 2020; 234: 4645–4658.

22.

Sahu

Pattanayak

Panda

. WEDM microdrilling of 316 L stainless steel orthopedic implant. Proc IMechE, Part C: J Mechanical Engineering Science 2020; 234: 3416–3435.

23.

Bose

Nandi

. Statistical and experimental investigation using a novel multi-objective optimization algorithm on a novel titanium hybrid composite developed by lens process. Proc IMechE, Part C: J Mechanical Engineering Science. Epub ahead of print 14 September 2020. DOI: 10.1177/0954406220959101.

24.

Haq

Marimuthu

Jeyapaul

. Multi response optimization of machining parameters of drilling Al/SiC metal matrix composite using grey relational analysis in the Taguchi method. Int J Adv Manuf Technol 2008; 37: 250–255.

25.

Babu

Marichamy

Ganesan

, et al. Processing of functionally graded aluminum composite brake disc and machining parameters optimization. Mater Today Proc 2019; 21: 563–567.

26.

Taşkesen

Kütükde

. Experimental investigation and multi-objective analysis on drilling of boron carbide reinforced metal matrix composites using grey relational analysis. Measurement 2014; 47: 321–330.

27.

Rajeswari

Amirthagadeswaran

. Experimental investigation of machinability characteristics and multi-response optimization of end milling in aluminium composites using RSM based grey relational analysis. Measurement 2017; 105: 78–86.

28.

Tosun

Muratoglu

. The drilling of an Al/SiCp metal-matrix composites. Part I: microstructure. Compos Sci Technol 2004; 64: 299–308.

29.

Cetin

Kabave Kilincarslan

. Effects of cutting fluids with nano-silver and borax additives on milling performance of aluminium alloys. J Manuf Process 2020; 50: 170–182.

30.

Aamir

Giasin

Tolouei-Rad

, et al. A review: drilling performance and hole quality of aluminium alloys for aerospace applications. J Mater Res Technol 2020; 9: 12484–12500.

31.

Karabulut

Karakoç

Çıtak

. Influence of B4C particle reinforcement on mechanical and machining properties of Al6061 /B4C composites. Compos Part B Eng 2016; 101: 87–98.

32.

Jadhav

Kakde

Patil

, et al. Effect of cutting parameters, point angle and reinforcement percentage on surface finish, in drilling of AL6061/Al2O3p MMC. Procedia Manuf 2018; 20: 2–11.

33.

Zou

Dang

Cai

, et al. Hole quality and tool wear when dry drilling of a new developed metal/composite co-cured material. Proc IMechE, Part B: J Engineering Manufacture 2020; 234: 980–992.

34.

Thirukkumaran

Menaka

Mukhopadhyay

, et al. A study on temperature rise, tool wear, and surface roughness during drilling of Al–5%SiC composite. Arab J Sci Eng 2020; 45: 5407–5419.

35.

Hakami

Pramanik

Basak

. Tool wear and surface quality of metal matrix composites due to machining: a review. Proc IMechE, Part B: J Engineering Manufacture 2017; 231: 739–752.

36.

Ramesh

Viswanathan

Ambika

. Measurement and optimization of surface roughness and tool wear via grey relational analysis, TOPSIS and RSA techniques. Measurement 2016; 78: 63–72.

37.

Taşkesen

Kütükde

. Optimization of the drilling parameters for the cutting forces in B4C-reinforced Al-7XXX-series alloys based on the Taguchi method. Mater Tehnol 2013; 47: 169–176.

38.

Senthilkumar

Tamizharasan

Anandakrishnan

. Experimental investigation and performance analysis of cemented carbide inserts of different geometries using Taguchi based grey relational analysis. Measurement 2014; 58: 520–536.

39.

Yaşar

. Thrust force modelling and surface roughness optimization in drilling of AA-7075: FEM and GRA. J Mech Sci Technol 2019; 33: 4771–4781.

40.

Übeyli

Balci

Sarikan

, et al. The ballistic performance of SiC-AA7075 functionally graded composite produced by powder metallurgy. Mater Des 2014; 56: 31–36.

41.

Erdemir

Canakcı

Varol

. Microstructural characterization and mechanical properties of functionally graded Al2024/SiC composites prepared by powder metallurgy techniques. Trans Nonferrous Met Soc China 2015; 25: 3569–3577.

42.

Gemi

Morkavuk

Köklü

, et al. An experimental study on the effects of various drill types on drilling performance of GFRP composite pipes and damage formation. Compos Part B Eng 2019; 172: 186–194.

43.

Neseli

. Optimization of process parameters with minimum thrust force and torque in drilling operation using Taguchi method. Adv Mech Eng 2014; 6: 925382.

44.

Melo

TFL

Ribeiro Filho

SLM

Arruda

ÉM

, et al. Analysis of the surface roughness, cutting efforts, and form errors in bore reaming of hardened steel using a statistical approach. Measurement 2019; 134: 845–854.

45.

Joy

Prakash

Krishnamoorthy

, et al. Experimental investigation and analysis of drilling in Grade 5 Titanium alloy (Ti-6Al-4V). Mater Today Proc 2019; 21: 335–339.

46.

Rajmohan

Palanikumar

Kathirvel

. Optimization of machining parameters in drilling hybrid aluminium metal matrix composites. Trans Nonferrous Met Soc China Engl Ed 2012; 22: 1286–1297.

47.

Wang

Zhang

, et al. Combined effects of various materials on tool wear in drilling of Ti/CFRP stacks. Proc IMechE, Part C: J Mechanical Engineering Science 2020; 234: 2750–2759.

48.

Sathish

Ramkumar

Geetha

. Drilling performances and wear characteristics of coated drill bits during drilling reinforced concrete. Int J Appl Ceram Technol 2019; 16: 357–366.

Investigation of drilling properties of AA7075/Al 2 O 3 functionally graded materials using gray relational analysis

Abstract

Keywords

Introduction

Material and methods

Fabrication of FGMs

Mechanical and microstructural analysis

Experimental setup

Experimental design for drilling

GRA

Results and discussions

Mechanical and microstructural properties

Thrust force optimization

Surface roughness optimizations

Mathematical modeling

Multiresponse optimization with GRA

Tool wear

Conclusions

Footnotes

Acknowledgements

Declaration of conflicting interests

Funding

ORCID iDs

References