Elsevier

Tribology International

Volume 151, November 2020, 106511
Tribology International

Tool wear and machined surface characteristics in side milling Ti6Al4V under dry and supercritical CO2 with MQL conditions

https://doi.org/10.1016/j.triboint.2020.106511Get rights and content

Highlights

  • scCO2 with MQL as an efficient coolant/lubricant is used in milling Ti6Al4V.

  • A theoretical model of VB with average prediction error 15.87% is established.

  • Tool wear causes the cutting torque to lag behind the rotation angle continuously.

  • Milled surface profile is characterized using continuous wavelet transform.

  • Performance of scCO2 with MQL is superior to scCO2 alone due to improved lubrication.

Abstract

Ti6Al4V alloy is a typical difficult-to-cut material. In order to improve its machinability and realize cleaner production, eco-friendly cooling/lubrication techniques are applied. Therefore, this study aims to investigate the tool wear, surface topography, cutting torque, and surface profile in side milling Ti6Al4V under four sustainable conditions, i.e., dry, supercritical carbon dioxide (scCO2), scCO2 with antifreeze water based minimum quantity lubrication (scCO2-WMQL), and scCO2 with oil-on-water based MQL (scCO2-OoWMQL) conditions. A theoretical model of flank wear width VB with average prediction error 15.87% is established. scCO2-OoWMQL reduces VB by 67.2% compared to scCO2 alone due to improved lubricity. Detailed characteristics of machined surface profile are investigated using continuous wavelet transform. The performance of scCO2-OoWMQL as a new sustainable and efficient cooling/lubrication technique is superior to scCO2 alone.

Introduction

Titanium alloy has good corrosion resistance, high specific strength and toughness even at extreme temperatures, so it gets increasing application in the aerospace industry and medical fields. However, the material spring back in the third deformation zone is large during the cutting process [1], which leads to severe friction and wear in the tool-workpiece interface. Moreover, the high chemical affinity and increased cutting temperature result from low thermal conductivity of titanium alloy increase the adhesion, diffusion and chemical wear of the tool. Apart from tool wear, severe work hardening and plastic deformation deteriorate the machined surface quality [2].

To solve the aforementioned problems, cooling/lubrication techniques are often applied in the processing of titanium alloys. The most common technique is wet cutting, also known as flood cooling. However, this technique requires pouring a large amount of metalworking fluids (MWFs) during the cutting process, which will increase production cost, pollute the environment, and harm the health of operators [3]. To achieve cleaner production, green cooling/lubrication techniques such as minimum quantity lubrication (MQL) and cryogenic cooling, as well as the combination of both, known as cryogenic MQL (CMQL), should be implemented.

The coolants used in cryogenic machining are usually liquid nitrogen (LN2) and supercritical carbon dioxide (scCO2). The scCO2-based cryogenic machining is a new sustainable green cooling/lubrication technique, which is proposed by Clarens et al. [4]. CO2 is a non-toxic gas at normal temperature and pressure. In the supercritical phase (above 7.38 MPa pressure and 31.2 °C temperature), scCO2 has excellent solubility for aliphatic and most aromatic hydrocarbons and thus can carry MWFs in solution [5]. After the scCO2 jet penetrates the cutting zone at a certain speed and pressure, it becomes a CO2 gas and a dry ice solid with a temperature of −78 °C due to the phase transformation [6], which significantly reduces the temperature of cutting zone and the surrounding environment. The lubricant film tends to rupture and evaporate due to high temperature and low penetration performance under MQL and wet cutting conditions (CCs). Therefore, the cooling performance of scCO2 is better than that of MQL and wet cutting. Compared with LN2, its advantages are as follows: the critical condition is easy to achieve (boiling point of LN2 is −196.6 °C); the production and storage cost of scCO2 is lower; scCO2 is less carcinogenic and has a low respiratory effect [7]; the effect of scCO2 on global warming is significantly smaller than that of LN2 [7]; it can dissolve lubricants, namely, scCO2-based MQL (scCO2-MQL), which has higher heat dissipation potential and efficiency due to evaporation of lubricants, and better lubrication performance than that of scCO2 or MQL alone [8]. Although the outlet temperature of scCO2 is not as low as that of LN2, LN2 will cause excessive cooling and increase the hardness of material [9].

Tool wear and surface quality of machined parts are very concerned in practical production. Liang et al. [10] comprehensively reviewed the influence of tool wear on the surface integrity of titanium and nickel alloys in recent years, including surface morphology, microstructural alterations, and mechanical properties. Stephenson et al. [5] found that compared with flood cooling, scCO2-MQL significantly reduced tool wear and improved material removal rate due to the improved lubricity. Khanna et al. [7] found that the tool wear in cryogenic machining with liquid CO2 was 44% and 68% lower than that in flood and MQL turning. Bagherzadeh and Budak [11] found that scCO2-MQL showed higher tool life, surface finish, and lower cutting temperature and tool chip contact length than scCO2 alone in turning Ti6Al4V. In milling 316L stainless steel, a reduction of 32% in tool average flank wear width VB was observed in scCO2-MQL condition over scCO2 alone [12]. Based on the Navier-Stokes equation and Reynolds lubrication equation, Wang and Clarens [13] established an analytical model for assessing MWFs penetration into the flank cutting zone in orthogonal cutting. This model demonstrated that scCO2-MQL penetrated the tool-workpiece completely, while conventional MWFs failed to penetrated the cutting zone fully, resulting in insufficient lubricant and cooling. Vazquez et al. [14] studied the surface geometric accuracy, surface quality, tool wear, and burr formation in micro-milling of titanium alloy under dry, jet application, and MQL CCs. Results showed that the best performance can be obtained in the MQL environment. Liang and Liu [15] found that tool wear pattern of rake face was adhesion-diffusion-abrasion wear, while the predominant wear mechanism of the flank face was adhesion-abrasion in turning Ti6Al4V. Tool wear induced machined surface defects including ploughing grooves, adhesive titanium, and surface burning, and surface cracks. Sharma and Meena [16] pointed out that during high-speed (500−1100 m/min) cutting, the β phase of the titanium alloy became unstable at high temperatures and tended to diffuse with Co matrix of tool which leads to chipping and attrition. Yang et al. [17] proposed an prediction model of flank wear in milling based on trajectory similarity and support vector regression.

MQL can combine not only with scCO2, but also with nano-sized solid lubricants to form nanofluid-MQL, which improves the thermal conductivity of MQL oil [18,19], and retains the oil particles to prevent the immediate release of the cutting oil from the cutting zone [20]. In turning process, nanofluid-MQL reinforced by multi-walled carbon nanotubes [19], or reinforced by hexagonal boron nitride (hBN) nanoparticles [20], obtained lower surface roughness, cutting temperature, and tool wear compared with pure-MQL environment. Şirin and Kıvak [18] found that the optimal concentration for the nanofluids was 0.50 vol%. The performance of 0.5 vol% hBN is better than Al2O3 nanoparticles owing to the higher heat coefficient and more spherical shape of hBN [21]. Yin et al. [22] analysed the power spectrum density (PSD) of the surface profile in milling under dry, wet, MQL, and Nanofluid MQL conditions, pointing out that the surface profile under Nanofluid MQL condition has the highest wave fitness, which means the highest surface quality.

It can be seen from the above literature that cryogenic machining can inhibit tool wear and improve surface finish. Whereas Cai et al. [23] found that the surface roughness Ra and cutting force in the scCO2 condition increased by 22% and 37.8% respectively, compared with dry cutting in side milling Ti6Al4V. This is because deep cooling hardens the material, resulting in a higher cutting force and thus intensive forced-vibration. Yıldırım et al. [24] also demonstrated that MQL is more effective than cryogenic machining in reducing tool wear in turning process.

Although there are some predictions models of flank wear based on data processing method, this method does not reflect the physical phenomenon and cutting mechanism in machining process. There is no research on the theoretical modelling of tool wear in side milling Ti6Al4V under sustainable green CCs, i.e. dry, scCO2, scCO2-WMQL, scCO2-OoWMQL, and no detailed analysis of the milled surface profile characteristics. In this paper, based on the tool-workpiece contact characteristics in the third deformation zone, a theoretical model of tool average flank wear width VB under various CCs is established. This model directly associates with tool wear mechanism. The tool wear mechanism and its influence on cutting torque and surface morphology are also analysed. Additionally, based on the continuous wavelet transform (CWT), the surface profile features are investigated in the cutting length-scale domain.

Section snippets

Contact length of tool-workpiece interface

Fig. 1 is a schematic diagram of the milling process. Based on the linear edge force model, the total milling force can be divided into cutting force components and edge force components, which result from the shear slip in the primary deformation zone and ploughing/friction in the third deformation zone (tool-chip interface) respectively. The differential milling forces in the tangential, dFt, radial, dFr, and axial, dFa, are expressed as follows [25]:dFtj(φ,z)=[Ktchj(φ,z)+Kte]dzdFrj(φ,z)=[Krch

Cooling/lubrication conditions

Four cooling/lubrication techniques are adopted in this experiment, i.e., dry, scCO2, scCO2-WMQL and scCO2-OoWMQL. The CMQL device (Cryolube eOoW type) is manufactured by Dongguan Armorine Machinery Manufacturing Technology Co.,Ltd (China). In the mixing chamber of CMQL generator, under the action of air jet at a certain pressure and speed, water molecules adsorb oil molecules to form oil-on-water droplets. As shown in Fig. 4a and b, steel pipes convey scCO2, while plastic pipes convey

Tool wear

Fig. 6 depicts the experimental and predicted values of VB under four CCs. The experimental value is the average of VB for all cutter teeth, and the error bar represents standard deviation. Optical micrographs of flank wear land for all teeth and CCs is shown in Fig. 7.

The average prediction error of this theoretical model is only 15.87%, indicating that the model is reliable. VB under four CCs in descending order is scCO2 > dry > scCO2-WMQL > scCO2-OoWMQL (286.7 > 135.9>121.5 > 94.0 μm), which

Conclusions

With excellent solubility and cooling performance, scCO2 with MQL as a new sustainable and efficient coolant/lubricant is used in milling Ti6Al4V. This study establishes a theoretical prediction model of VB, and investigates the tool wear mechanism, its corresponding surface topography, cutting torque, and milled surface profile's CWT scalogram under dry, scCO2, scCO2-WMQL and scCO2-OoWMQL CCs. The following conclusions are drawn:

  • VB is positively correlated with Kre. Despite the excellent

Limitation and future scope

The reasonable arrangement of the nozzle can ensure the penetration of coolants/lubricants into the high-temperature cutting zone, thereby achieving optimal cooling/lubrication performance. The nozzle arrangement includes the target distance between nozzle and tool tip, and the angle between spray direction and feed direction. In this paper, the influence of nozzle arrangement on the machinability of Ti6Al4V is not studied. Moreover, the effect of cooling/lubrication conditions on residual

CRediT authorship contribution statement

Qinglong An: Conceptualization, Investigation, Resources, Funding acquisition. Chongyan Cai: Methodology, Investigation, Formal analysis, Writing - original draft, Writing - review & editing. Fan Zou: Formal analysis, Data curation. Xu Liang: Data curation, Software. Ming Chen: Supervision, Project administration.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

The work is supported by National Key R&D Program of China (2018YFB2002200).

References (32)

Cited by (85)

View all citing articles on Scopus
View full text