Experimental investigation and optimization of hybrid turning of Ti6Al7Nb alloy under nanofluid base

Abstract

Keywords: Ultrasonic vibration assisted machining; minimum quantity lubrication; machining response; titanium alloy; TOPSIS.

Introduction

Titanium alloys find extensive utilization in the field of biomedical devices due to their favorable characteristics, including low density, outstanding biocompatibility, exceptional resistance to corrosion, and impressive mechanical properties [1–3]. This is especially notable in load-bearing applications like orthopedic implants [3, 4]. Among titanium alloys, Ti-6Al-7Nb has been successfully used in a range of clinical applications, including hip and knee implants, dental implants, and spinal implants and researchers have reported positive clinical outcomes with this alloy, including reduced complications and improved patient comfort [5–8]. Numerous studies have demonstrated that the alloy is well-tolerated by the human body and exhibits well

osteoblast response when implanted [9, 10]. The elasticity modulus of Ti-6Al-7Nb is lower than that of Ti-6Al-4V [11, 12]. As a result of this, Ti-6Al-7Nb alloy facilitates better integration with bone and potentially leads to less stress shielding in load-bearing areas [13]. Moreover, Ti-6Al-4V alloy comprisal vanadium, which possesses toxic properties, limits the use of this alloy in biomedical applications [14, 15]. Therefore, the Ti-6Al-7Nb alloy stands apart from other titanium alloys in biomedical applications. On the contrary, Ti-6Al-7Nb alloy are often categorized as difficult-to-cut material, primarily due to their inherent properties like high strength and hardness, low thermal conductivity etc. [16]. These properties lead to rapid tool wear and plastic deformation of cutting tools reported by researchers [17, 18]. For instance, Risco-Alfonso et al. [19] conducted turning exper-

*Corresponding author. *E-mail address: erkinnduman@gmail.com Published by Yıldız Technical University Press, İstanbul, Türkiye This is an open access article under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

J Adv Manuf Eng, Vol. 4, Issue. 2, pp. 35–45, December, 2023

Uva-Nmql

MQL: Minimum quantity lubrication; UVA-MQL: Ultrasonic vibration assisted-MQL; NMQL: Nanofluid MQL that is containing 0.8% graphene by weight; UVA-NMQL: Ultrasonic vibration assisted – nanofluid added MQL

iments on Ti-6Al-7Nb alloy using ceramic tools under dry cutting conditions. Their results showed that at the cutting speed of 200 m/min, the tool life was limited to only 6 minutes. While such tool life may be acceptable for many steel materials, it is notably high for machining this particular alloy. Carvalho et al. [17] focused on the dry turning of the Ti-6Al-7Nb alloy. In all the chosen cutting conditions, the chips did not break, and instead, they obtained unsuitable segmented and long helical ribbon chips. As a consequence of this research, several strategies such as micro cutting, and different cooling and lubrication techniques have been proposed to improve the machinability of Ti-6Al-7Nb alloys including minimum quantity lubrication (MQL) [20], cryogenic and flood coolants [13], micro cutting [21]. Among these strategies, MQL has arisen as a favorable substitute for conventional flood cooling owing to an environmentally conscious method that involves the application of a very small amount of mineral or vegetable oil in mist form, delivered by a compressed air stream (typically 5–7 bar), directly to the cutting zone [22]. MQL has demonstrated remarkable performance in turning, milling, and grinding operations by penetrating the cutting zone and providing essential lubrication [23]. The implementation of this method has greatly aided in enhancing the chip forming process of numerous

engineering alloys. For instance, in steel machining, Jagatheesan et al. [24] revealed that comparison to traditional dry cutting and flood machining, the MQL method has achieved a reduction of approximately 20% in cutting forces and 12% in cutting temperature, thereby yielding an improved surface quality. Similarly, Kannan et al. [25] demonstrated the impact of the MQL method on reducing tool wear and cutting forces in the machining of aluminum alloy and its composites. Furthermore, Gong et al. [26] conducted a study on the impact of the MQL method on surface integrity during the machining of Inconel 718 alloy. The MQL method diminished surface defects, and creates a smoother surface compared to dry and flood cutting, with fewer marks and adhered particles to it. Nowadays, there has been a growing adoption of nanoparticle-added super lubricants to improve the effectiveness of MQL in machining [27, 28]. Multiple types of nanoparticles, such as Al2O3 [29], MoS2 [30], CuO [31], zinc oxide (ZnO) [32], carbon nanotube [33], and graphene [26] etc. have been employed for the purpose of nanofluid preparation and subsequent examination of their effects. Therefore, super lubricity has garnered significant global interest at a time when humanity is facing a critical energy crisis [34]. Among these, graphene is one

J Adv Manuf Eng, Vol. 4, Issue. 2, pp. 35–45, December, 2023

Figure 1. Flow chart of experimental methodology. of the super lubricants and it is an extremely thin material at the atomic level, possessing a low surface energy that has the potential to reduce both friction and adhesion [35]. As a result, it has been increasingly employed in machining applications as both a coolant and a lubricant. Lately, nontraditional machining methods have been employed in addition to conventional cooling and lubrication conditions to enhance the machinability of difficult to cut materials. One of these methods is ultrasonic vibration assisted (UVA) machining. Vibration-assisted machining involves the application of vibrations to either the cutting tool or the workpiece, typically at specific frequencies and amplitudes. This method offers distinct advantages, particularly when dealing with difficult-to-machine materials, in comparison to conventional machining approaches. It effectively mitigates undesirable noise, reduces tool wear, and enhances the surface finish of the machined workpiece [36]. Koshimizu et al. [37] investigated ultrasonic vibration machining of Ti-6Al-4V alloy. They found that applying ultrasonic vibration to the tool tip reduced cutting force and improved tool wear and surface roughness [37]. Kandi et al. [38] investigated ultrasonic vibration-assisted turning and discovered reductions in cutting forces and surface roughness for Ti-6Al-4V alloy. Furthermore, studies have shown that combining MQL and ultrasonic-assisted machining improves the performance of difficult-to-cut materials. For instance, Airao et al. [39] investigated the use of hybrid conditions involving MQL, cryogenic cooling and lubrication in the ultrasonic turning of Inconel 718 alloy. The results demonstrated that this hybrid machining approach significantly reduced flank wear by 32–53%, power consumption by 11–40%, and power consumption by 5–31% [9].

In light of the studies mentioned in the literature review, the need to try new strategies to improve the machining of Ti-6Al-7Nb alloy exists. The main objective of this work is to investigate the machinability of the Ti-6Al7Nb alloy using the MQL (Minimum Quantity Lubrication) method and to gain a better understanding of the impact of applying ultrasonic vibration and MQL with the addition of nanographene (NMQL) to the cutting tool on machining performance. In the second phase of this study, in order to determine the optimum cutting parameters for machining this alloy using the selected methods, the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) method was employed to determine the optimum cutting parameters.

Materials And Methods

In this work, biomedical grade Ti-6Al-7Nb alloy work material was used in order to perform machining experiments. The work material has 883 MPa yield strength, and 946 MPa ultimate tensile strength. Initial diameter of 60 mm and a length of 130 mm as indicated in Figure 1. GOODWAY GA-230 CNC lathe with a main power of 11 kW and a spindle speed up to 4200 rpm was utilized to carry out oblique cutting experiments. The machining length and depth of cut were kept constant at 15 mm and 1 mm respectively. The cutting tool insert utilized was of the uncoated tungsten carbide type Sandvik SCMT 120408-KM H13A with SSSCL 2525 M12 tool holder. The plan used in machining experiments is given in Table 1. In ultrasonic tests, an ultrasonic horn, ultrasonic generator and power unit were employed with supply voltage

J Adv Manuf Eng, Vol. 4, Issue. 2, pp. 35–45, December, 2023

Figure 2. Variation of resultant cutting force vs. cutting conditions at different feed (a) 0.1 mm/rev and (b) 0.14 mm/rev. with 230 V/50-60 Hz, and the frequency of the vibration was about 20 kHz detected by the oscilloscope-indicator, while the amplitude was measured 20 μm. Additionally, vibration was given to the cutting tool in a direction that was transverse to the main cutting force axis. An MQL device is employed for delivering commercial vegetable cutting fluid to the cutting zone. It operated with a flow rate of 30 mL per hour and maintained a pressure of 5 bar. The MQL nozzle had a diameter of 1 mm, and it was positioned at an angle of 30º. At stage of the nanofluid, nano graphene particles were used with a particle size of 4–8 µm, gray-colored, had a purity of 99%, a surface area of 110–130 m2/g, a thickness of 5–10 nm, and a density of 2 g/cm3. The particles were spread into a glass container and subjected to a 75-minute moisture removal and drying process in an oven at a temperature of 120 °C. The dried nano particles were weighed on a precision balance and mixed with vegetable cutting oil at a ratio of 0.8% by weight. To enhance the homogeneity of the nGP nanofluid and limit clumping and sedimentation times, SDS (Sodium Dodecyl Sulfate) was added at a ratio of 0.1% by weight of graphene. Finally, the prepared mixtures were stirred for 60 minutes at 5000 rpm using a digital homogenizer. The procedure for preparing the nanofluids is shown in Figure 1. From the outputs in this study, cutting forces and cutting temperature were measured using the Kistler 9257BA model dynamometer and the Optris – CT laser 3MH1 model double laser contactless temperature measurement respectively. Besides, A Mitutoyo SJ-210 surface roughness tester was used to measure machined surface roughness values. The flow chart of experimental methodology is depicted in Figure 1.

Results And Discussion

Analysis of Resultant Force The cutting force is considered a vital aspect of the chip removal process. Various factors impact these cutting forces, such as cutting speed, feed, depth of cut, cutting tool, and the type of cutting fluid in use. Resultant force is calculated by utilizing Eq.1 [40] FR2=F2x+F2y+F2z(1) Here, FR is the resultant force, Fx, Fy and Fz represent radial cutting force, main cutting force, and feed force re-

spectively [40]. Calculated resultant force results are depicted at the feed rate of 0.1 mm/rev in Figure 2a and 0.14 mm/ rev in Figure 2b. The obtained findings have demonstrated that the combination of NMQL (nanographene-added MQL) with ultrasonic vibration-assisted machining has enabled a significant reduction in cutting forces for all selected cutting conditions. For instance, at a cutting speed of 70 m/min and the feed rate of 0.1 mm/rev, the resultant force recorded in the MQL was 223 N, whereas, with the application of UVA-NMQL, the force was measured to be approximately 24% lower at 170 N. When compared to MQL machining, the utilization of UVA-NMQL resulted in about 12% reduction in the resultant force at the cutting speed of 130 m/min and the feed of 0.14 mm/rev. Compared to cutting speed, feed rate has a more significant impact on forces. An increase in feed rate has been observed to lead to an increasing trend in cutting forces for all cutting conditions. For low cutting speed (at 70 m/min) in continuous turning, when feed increased from 0.1 mm/rev to 0.14 mm/rev, the resultant force increased from 225 N to 264 N in MQL and 192 N to 245 N in NMQL. Apparently, the NMQL method has led to lower cutting forces compared to the MQL method. The main reason for this phenomenon is attributed to tremendous properties of nanographene like atomic-level thinness and remarkably low surface energy, offering the promise of decreasing both friction and adhesion [35]. The cutting temperature results shown in Figure 3 qualitatively supports these findings. However, in ultrasonic vibration assisted machining, recorded forces are 210 N to 240 N in UVA-MQL and 171 N up to 228 N in UVA-NMQL for selected feed 0.1 mm/rev and 0.14 mm/ rev, respectively. This outcome is probably an effect of the intermittent cutting process introduced through ultrasonic vibration-assisted machining. In conventional continuous turning (CT), the tool maintains constant contact with the workpiece, resulting in relatively elevated resultant forces. Conversely, intermittent cutting features diminish the friction between the tool and workpiece, thereby leading to a reduction in cutting forces observed in ultrasonic vibration assisted turning (UVAT). It's worth noting that when the UVAM and both MQL and NMQL methods are used simultaneously, the lubrication and cooling performance of MQL can be further improved. This phenomenon has

J Adv Manuf Eng, Vol. 4, Issue. 2, pp. 35–45, December, 2023

Figure 3. Cutting temperature variation for different cutting conditions and feed (a) 0.1 mm/rev; (b) 0.14 mm/rev. been explained due to the separate-type cutting mechanism of UVAM and cavitation phenomenon of the resonant workpiece [41]. It has been reported that using UVA and MQL simultaneously has a coupling effect [42]. When high temperature thermal stress and poor friction qualities cause microcracks and friction traces to emerge on the tool surface, UVAMQL can effectively alleviate these problems reported by Ni et al. [43]. Cutting Temperature Low thermal conductivity (6.6–6.8 W/m.K) of titanium alloys leads to elevated heat generation during the machining of these alloys due to the plastic deformation and friction effects at the tool-chip interface and shear plane [40, 44, 45]. Therefore, cutting temperature is a critical factor for understanding the machining performance of titanium alloys. Figure 3 displays cutting temperature variation for different cutting conditions. In this study, it was observed that the implementation of MQL resulted in the attainment of the highest recorded temperature of 159 °C. This temperature was achieved under the cutting conditions of a speed of 130 m/min and a feed rate of 0.14 mm/rev. However, thanks to the employment of NMQL the cutting temperature decreased by approximately 10%, measuring 143 °C. However, at a relatively lower cutting speed of 70 m/min, a more significant reduction of approximately 17% in temperature was achieved with NMQL. These findings can be attributed to its high thermal conductivity and lubrication properties resulting from the presence of graphene. Moreover, compressed air improves cooling and nanofluid lubricates machining interfaces by providing rolling effect and protective film [30]. The friction coefficient, thermal conductivity, and dynamic viscosity of cutting oil are also critical parameters for heat transfer. These parameters are predominantly influenced by the tribological and thermophysical characteristics of the prepared nanofluid. However, formation of a tribofilm in cutting zone is attributed not only to lubrication but also to the application of air at a specific pressure in the MQL method. Obikawa et al. [46] tested the effectiveness of the MQL system at air pressure values ranging from 3 to

7 bar. They found that an increase in air pressure reduced tool wear. On the other hand, Li and Liang [47] focused on convective heat transfer separately for air and oil in the MQL method. In a study conducted by Kurgin et al. [48], the effect of different air pressure and oil flow rates on the convective heat transfer coefficient in the MQL technique was investigated. Another study by Gong et al. [26] found that the specific heat capacities of cutting oil and graphene nanoparticles suspended in NMQL techniques varied. Added graphene nanoparticle increased the MQL-specific heat capacity by 11%. Based on the literature review, the MQL method used in this study was conducted with a vegetable-based cutting oil with a thermal conductivity of 0.5–1 W/m.K [18]. In contrast, graphene is produced from synthetic graphite powder [19] and graphene nanoparticles can exhibit an impressive thermal conductivity of up to 3000–5500 W/m.K [20, 21]. Consequently, graphene added nanofluid based MQL (NMQL) is improve heat transfer. At a high cutting speed of 130 m/min and feed of 0.14 mm/rev, the effectiveness of NMQL was found to diminish. This can be attributed to insufficient time for cooling and diffusion processes. On the other hand, the lowest cutting temperature was achieved under the UVA-NMQL condition. The application of the NMQL method in conjunction with ultrasonic vibration significantly reduced the cutting temperature. As a result of the intermittent cutting process between the tool-workpiece pair in ultrasonic vibration-assisted machining, long-term extrusion between the workpiece and tool free surface, as well as chip removal on the tool free surface, is interrupted. This situation leads to a reduction in plastic deformation in the chip deformation zone, resulting in a decrease in cutting temperature [16]. It is understood that the ultrasonic vibration-assisted machining process provides a significant decrease in cutting temperature as a result of the intermittent cutting process and the separation of the tool-workpiece pair. Surface Roughness Essential surface roughness values within nominal limits are expected in biomedical applications in order to ensure conformity between tissue and implants [49–51].

J Adv Manuf Eng, Vol. 4, Issue. 2, pp. 35–45, December, 2023

Figure 4. Surface roughness variation for different cutting conditions and feed (a) 0.1 mm/rev; (b) 0.14 mm/rev. Moreover, machining process generates smoother surfaces and thereby corrosion resistance, fatigue life etc. characteristics have been influenced at the end of the process [52]. Therefore, average surface roughness values of processed samples subjected to CT and UVA machining under variable cutting parameters and conditions have been examined. The measured surface roughness results are presented in Figure 4. It can be seen that the surface roughness values obtained are below 1 μm under all cutting speeds and cutting conditions. A higher roughness peak was measured in MQL machining at the feed of 0.1 mm/rev, whereas the lowest roughness values have been obtained with nano graphene particle added MQL and ultrasonic vibration-assisted machining (UVA-NMQL). In comparison with MQL, it was evident that decreasing of surface roughness by approximately 30%, 44% and 49% under UVA-MQL, NMQL and UVA-NMQL, respectively when the cutting speed was 100 m/min and the feed rate was 0.1 mm/rev. The ability of graphene nanoparticles to reduce the coefficient of friction is the reason for this phenomenon [53]. Moreover, in MQL machining, at the cutting speed of 130 m/min and feed rate of 0.1 mm/ rev, the application of ultrasonic vibration to the cutting tool (UVA-MQL) reduced the surface roughness by 32%. Ni et al. [43] conducted experiments using CT, MQL, and MQLUVAM methods separately in the milling process of the Ti-6Al-4V alloy. They reported that the simultaneous application of the UVAM and MQL methods resulted in a synergistic effect, which considerably reduced cutter tool wear and minimized the formation of microcracks in the tool. As a result, the surface quality of the UVAMQL method was significantly improved [43]. It is reasonable to expect that using the UVAM and MQL methods simultaneously will produce excellent microtextured surfaces. However, when a feed rate of 0.14 mm/rev was chosen, this positive effect dramatically decreased, and the impact of ultrasonic vibration on reducing surface roughness was approximately 6%. On the other hand, all measured surface roughness values are smaller than the predicted average roughness value according to Equation (2) [40], which were calculated as 0.39 µm, and 0.77 µm for 0.1 mm/rev and 0.14 mm/rev feed, respectively.

(2) where ƒis the feed and rc is the depicted corner radius of the tool. Multi-Criteria Decision Making with TOPSIS In the realm of manufacturing and machining, achieving optimal performance in machining processes is paramount for ensuring product quality, energy efficiency, and cost-effectiveness [54]. The quest for excellence in machining operations necessitates a meticulous analysis of process parameters to attain the finest balance between multiple conflicting objectives such as minimizing cutting force and cutting temperature, maximizing material removal rates, and ensuring surface finish quality. To address these multifaceted concerns, researchers and engineers have turned to Multi-Criteria Decision-Making (MCDM) methods as a powerful approach for optimizing the machining process [55]. Among the various MCDM techniques, the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) method stands out as a robust and versatile tool that can be adeptly applied to the domain of machining optimization. Prior to assessing the proximity of the outcomes to the ideal outcome in the TOPSIS methodology, it is necessary to compute the weighted efficiency distributions of the respective activities. The distribution of efficacy can be assigned uniformly across all outcomes or determined by the researcher through their expertise or mathematical equations. In this study, the equations presented by Saatçi et al. [56] were used to calculate the efficiency distributions. The method used to calculate the weighted efficiency distributions here is the entropy method. The distributions of the results' weighted efficiencies were determined as 0.228, 0.268, 0.234, and 0.27 for cutting force, surface roughness, cutting temperature, and material removal rate, respectively. Following the computation of the outcomes' weighted efficiency distributions, the TOPSIS decision-making method operates. In the presented study, while the material removal rate was maximized, the cutting force, cutting temperature, and surface roughness were minimized and sorted according to the closeness to the ideal solution. The TOPSIS method proposed by Ous-

J Adv Manuf Eng, Vol. 4, Issue. 2, pp. 35–45, December, 2023

Figure 5. Optimal results selected by the TOPSIS. sama et al. [57] and applied in this study includes the stages of normalization of the decision matrix, calculation of the weighted normalized decision matrix, calculation of the negative and positive values of the ideal solutions, calculation of the distance between each alternative solution and the positive and negative ideal solution, and ranking the solutions in descending order. Consequently, the optimal results obtained through the TOPSIS method are presented in Figure 5. The best result, which reduces cutting force and cutting temperature and increases material removal rate and surface quality, was obtained at the cutting speed of 130 m/min and at the feed rate of 0.1 mm/rev under UVA-NMQL cutting conditions. This was followed by using a cutting speed of 130 m/min and the feed rate value of 0.1 mm/rev under the NMQL cutting condition. It is clear from the TOPSIS decision-making process that a low feed value is the best option. A similar result was presented by Saatçi et al. [56] in the orthogonal turning of AISI 310S stainless steel to obtain minimum cutting force, machining cost, and carbon emission in the selection of machining parameters with the TOPSIS method. Additionally, as can be seen from Figure 5, it can be seen that high cutting speed is the best option when evaluated together with the weighted efficiencies obtained in terms of cutting force, cutting temperature, metal removal rate, and surface roughness. Actually, reducing the cutting speed causes a decrease in cutting tool wear, resulting in a decrease in machining costs and carbon emissions resulting from the use of cutting tools [58]. For example, Jawaid et al. [59] determined that the extent of flank wear and cutting edge deformation of the cutting tool increased as the cutting speed increased in the turning of Ti-6246 titanium alloy. However, in terms of sustainable turning,

decreasing the cutting speed increases the machining time, thus increasing the energy consumed, resulting in an increase in the total machining cost and carbon emissions [60]. As a result, increasing the cutting speed, reducing the feed rate, and using ultrasonic vibration-assisted machining offers the best results in terms of machining response and sustainability.

Conclusions

In order to enhance the efficiency of the MQL method employed to achieve sustainable manufacturing objectives, this study investigates the machining performance of the Ti-6Al-7Nb biomedical titanium alloy. It does so by utilizing nano-fluids prepared with the addition of graphene nanoparticles to MQL oil, alongside energy-efficient and power-specific ultrasonic vibration-assisted machining techniques. The analysis focuses on output parameters such as cutting force, cutting temperature, material removal rate and surface roughness. The specific findings obtained from this investigation are outlined below: 1) In experiments conducted using the conventional turning (CT) method, the NMQL method containing 0.8% by weight of graphene has reduced resultant forces by a range of 6% to 14% compared to vegetable-based cutting oil assisted MQL. 2) UVA-NMQL hybrid method has resulted in a reduction in cutting forces by approximately 9% to 19% compared to the UVA-MQL method. Besides, when compared to the traditional MQL method, which generated a resultant force value of 223 N, the UVA-NMQL method reduced the cutting force by approximately 23%, bringing it down to 171 N, which is considered significant for the machining of this alloy.

J Adv Manuf Eng, Vol. 4, Issue. 2, pp. 35–45, December, 2023

3) UVA-NMQL method exhibits the lowest cutting temperature. In this study, under high cutting conditions, which are considered to be Vc=130 m/min and f=0.14 mm/rev, the UVA-NMQL method has successfully reduced the cutting temperature by approximately 15%, bringing it down from 159 °C to 136 °C. 4) The hybrid method uses intermittent cutting and the nanofluid containing graphene, which falls into the superlubricant category, in the UVA-NMQL method, resulted in low cutting forces and temperatures, contributing positively to surface roughness. As an example, in the case of the MQL method at the cutting speed of 130 m/min. and the feed rate of 0.1 mm/rev., the average surface roughness, which was initially 0.7 µm, was reduced by approximately 47% to a measurement of 0.37 µm when the UVA-NMQL method was implemented. This trend was consistently observed across all selected cutting parameters. 5) The optimum cutting parameters selected by TOPSIS are UVA-NMQL cutting conditions, 130 m/min. of cutting speed, and 0.1 mm/rev of feed rate value, when cutting force, surface roughness, cutting temperature, and material removal rate are evaluated together. In future studies, the effect of the hybrid nanofluid MQL methods on the machining performance of the Ti6Al-7Nb biomedical titanium alloy will be examined. In addition, it is planned to conduct a sustainability assessment in terms of carbon emissions, consumed energy, and processing costs.

Share and Cite