This paper studies the effect of various lamellar-type solid lubricants (graphite and hBN) that can be mixed into a lubricant to potentially improve the machinability of minimum quantity lubrication (MQL) machining. To examine this, the solid lubricants are classified into particles and platelets based on their aspect ratios as well as their respective sizes. In particular, the particles are classified into microparticles and nanoparticles based on their dimensions (average radius), while the platelets were classified, based on their average thickness, into two types: the “microplatelets” if the thickness is typically up to few tens of microns and the “nanoplatelets” if the thickness is well below a tenth of a micron (even down to few nanometers). Our previous work has shown that the mixture of an extremely small amount (about 0.1 wt. %) of the graphitic nanoplatelets and vegetable oil immensely enhanced the machinability of MQL machining. In this paper, many lubricants, each mixed with a particular variety of nano- or micro-platelets or one type of nanoparticles, were studied to reveal the effect of each solid lubricant on MQL machining. Prior to the MQL machining experiment, the tribological test was conducted to show that the nanoplatelets are overall more effective than the microplatelets and nanoparticles in minimizing wear despite of no significant difference in friction compared to pure vegetable oil. Consequently, the MQL ball-milling experiment was conducted with AISI 1045 steel yielding a similar trend. Surprisingly, the oil mixtures with the microplatelets increased flank wear, even compared to the pure oil lubricant when the tools with the smooth surface were used. Thus, the nanoscale thickness of these platelets is a critical requirement for the solid lubricants in enhancing the MQL machining process. However, maintaining the nanoscale thickness is not critical with the tools with the rough surfaces in enhancing the MQL process. Therefore, it is concluded that finding an optimum solid lubricant depends on not only the characteristics (material as well as morphology) of solid lubricants but also the characteristic of tool surface.
Traditional flood cooling is widely used in industrial machining operations. The primary purposes of metalworking fluid (MWF) in flood cooling are to dissipate the heat rapidly, to reduce friction between tool and chip, and to carry the chips away from the cutting zone. Consequently, the reduction in force, power consumption, improvement in tool life, and surface finish are apparent  with flood cooling. However, it suffers from its associated maintenance/disposal cost (7–17%), occupational hazard, and environmental impact . Dry machining offers an alternate solution by completely eliminating MWF. However, poor surface quality, accelerated tool wear, and a high concentration of the airborne particles on the shop floor hampers its applications [3,4].
MQL provides an ideal compromise between dry and flood cooling conditions. MQL machining refers to one class of machining processes where the lubricant use is limited to only a small amount—typically a flow rate of 5–500 ml/hr [5,6]. This amount is at least about three to four orders of magnitude lower than the amount commonly used in flood cooling, which dispenses up to 60 L/h . The most common lubricants used in MQL machining are mineral oil [5,8–11], emulsions (soluble or synthetic fluid) [10,12,13], synthetic oils, esters [10,13], neat oil [14,15], and vegetable oil [5,13,16–19]. Some researchers have used alcohol [5,20] and de-ionized water [21,22] as well. In general, vegetable oil is favored because of the reduction in tool wear [23,24], cutting force , and surface roughness [17,26,27]. In particular, Tai et al.  evaluated the various MQL liquid lubricants and determined the lubricant parameters influencing the machinability. More importantly, the biodegradability of the vegetable oils makes them safer for the workers and the environment in comparison to other lubricants. MQL in general offers many benefits such as the diminution of the environmental and occupational hazards and the reduction of maintenance and disposal cost. However, MQL does not provide the ample cooling because of the extremely small amount of a liquid lubricant. Basically, MQL provides essential lubricant function, which restricts its applications to aggressive cutting conditions (e.g., high cutting speed).
To improve the performance and expand the range of MQL applications, many spray configurations and cutting parameters are proposed and investigated, while others have studied to enhance the lubricity of a liquid lubricant by adding solid lubricants [5,23,24,28,29]. Few studies concentrated on the enhancement of the nozzle design. Oliveira et al.  used an extra compressed air nozzle to clean up the chips and the MQL oil remnant from the grinding wheel before applying the MQL oil on each grinding cycle on AISI 4340 steel. The MQL with the compressed air applied with the incidence angle of 30 deg shows better surface finish compared to both conventional flood cooling and traditional MQL (without compressed air). Aoyama et al.  developed a new droplet supply system with four compressed air inlets in order to reduce the amount of oil used in MQL (only 0.1 mg/m3 compared to the average 15 mg/m3 used in typical MQL applications) while keeping the performance of this system is comparable to those of conventional MQL. Despite the extremely small amount of oil used, the flank wear was significantly improved (up to 30%) compared to dry machining of Ti–6Al–4 V. Nath et al.  developed a new atomization-based cutting fluid spray system to improve the productivity in machining titanium alloys. Itoigawa et al.  designed the MQL nozzle to generate oil-film on water (OoW) droplets in order to provide cooling. Even though the results of friction test with OoW droplet were similar with those of conventional MQL, the cutting forces in machining of AlSi5 alloy showed remarkable reduction with synthetic ester but not with rapeseed oil. Some researchers examined solid lubricants as potential additives to MQL lubricant in order to extend the applicability of MQL. The solid lubricants typically used as dry powders or coated materials include molybdenum disulfide (MoS2), graphite, hexagonal boron nitride (hBN), and polytetrafluoro-ethylene (PTFE) . The main advantage of solid lubricants is the preservation of lubricity even in extreme pressure and temperature prevalent in many machining applications. Shen et al.  used up to 20 wt. % of MoS2 nanoparticle with the particle diameter less than 100 nm to enhance the grinding process. Park et al.  showed that by using a very small amount (0.1 wt. %) of exfoliated graphite nanoplatelets (xGnP) made by XG Science, Inc. (Lansing, MI) reduced flank wear up to 50% in MQL ball milling of 1045 steel. Alberts et al.  found that 1 wt. % was the optimal concentration of xGnP for the grinding process of hardened tool steel.
The xGnPs with a lamellar crystal structure enable each layer to slide easily against the adjacent layers to provide the lubricity. Our previous work [23,24] used the nanoplatelets of graphite and hBN, known to be stable in the air up to 500 °C  and 1000 °C , respectively. When a small amount of these platelets mixed in MQL oil is used as the lubricant for MQL machining, they showed the impressive performance in extending tool life. However, making the nanoplatelets requires more sophisticated processing techniques. Therefore, this paper explored if more economical microplatelets solid lubricants can be used by making experimental comparison between particles and platelets as well as nano- and micro-platelets with a series of tribometer tests and MQL-ball milling. Furthermore, to maximize the performance of MQL, the MQL experiment must be conducted under the optimal MQL spray conditions including spray angle. The optimal MQL condition was adopted throughout this experimental work.
The effectiveness of MQL is directly dependent on the application angle of the nozzle. Figure 1 shows the nozzle orientations described by two Euler angles (yaw and pitch) in a ball-milling experiment. The experiment is set up for the rotating ball mill to travel along the X-axis shown in Fig. 1. In this setup, the pitch angle is defined as an elevated angle of the nozzle from the XY plane, while the yaw angle is the angle between the feed direction (X-axis on the work material) and nozzle spray direction. Many researchers [23,38–43] have worked on this topic and Table 1 summarized their findings in the optimal spray angles (both yaw and pitch angles if possible) and other spray parameters for external MQL systems based on the pitch and yaw angles defined in Fig. 1.
Several works were published to evaluate the effectiveness of MQL applications by mixing a small amount of solid lubricant into liquid lubricant. The solid lubricants were classified based on their structure such as polymers (i.e., PTFE and fluorocarbon resin PTFE (CF2)), lamellar solid (i.e., graphite, hBN, molybdenum sulfide (MoS2)), fluorides (i.e., LiF, CaF2, BaF2, and NaF), soft metals (Cu and Pb), and oxide solid lubricants (i.e., PbO, CuO, and B2O3) . In particular, the lamellar solid lubricants include graphite, hBN, boric acid, and the transition-metal dichalcogenides MX2 (M is molybdenum, tungsten, or niobium and X is sulfur, selenium, or tellurium, e.g., MoS2, WS2), and monochalcogenides (e.g., GaSe and GaS) . MoS2, which works best in a vacuum or dry condition, dissociates at the relatively low temperature (∼300 °C ) in a typical atmosphere condition with moisture and oxygen . As shown by Shen et al. , the flow rate of 5 ml/min of vegetable oil and paraffin oil mixed with 20 wt. % of MoS2 improved the tool life by up to 35% when compared with the traditional MQL with no solid lubricant. Srikant et al.  added CuO in the MQL turning process of AISI 1040 steel. They claimed that the mixture has improved the thermal conductivity and enhanced heat transfer. In Alberts et al. , the effect of two distinct diameters (average 1 and 15 μm) was investigated by mixing 2 wt. % xGnP into two lubricants, isopropyl alcohol and a common cutting semisynthetic, and water-based emulsion (Trim SC200) in MQL grinding. The results showed that the larger 15 μm-diameter xGnPs performed slightly better in terms of surface roughness, grinding force, and specific energy.
Ramana et al.  introduced the spherical nanocrystalline boric acid (the sizes of 50, 60, 80 nm and 0.538 μm) into canola oil to MQL turning of AISI 1040 steel heat-treated to 29 HRC. The significant improvements in cutting temperature, cutting forces, and surface quality were observed with the largest particle size (0.538 μm). They concluded that the nanocrystalline boric acid was not preferred in the machining of hardened steel. Kalita et al.  conducted the MQL grinding process on EN24 steel and cast iron with 2 and 8 wt. % mixed MoS2 nanoparticles (40–70 nm) and microparticles (3–5 μm) in paraffin and soy bean oil. They showed that the 8 wt. % mixture was better and the nanoparticles exhibited superior performance than the microparticles, resulting in up to 30% and 50% reductions in the friction coefficient and grinding ratio, respectively. The increase in the particle content reduced the friction, energy consumption, and cutting forces. The paraffin oil showed slightly better performance in enhancing the wheel life (G-ratio) than the soybean oil on the cast iron and the opposite was true for steel. With an MQL ball milling of AISI 1045 steel , the optimum xGnP concentration (0.1 wt. %) exists, beyond which tool wear did not improve. However, with the hBN nanoplatelets, Nguyen et al.  found that a higher concentration (0.5 wt. %) was better in improving tool life. Krishna et al.  conducted the MQL turning of AISI 1040 steel with 0.25, 0.5, and 1.0 wt. % of boric acid nanoparticles (50 nm) dispersed in SAE-40 and coconut oil. The performance in terms of the cutting temperature, surface roughness, and wear was improved with the 0.5 wt. % concentration with both oils. The results also showed the better performance with coconut oil.
Another approach is to utilize the bearing effect by the spherical particles mixed with lubricant. To produce better surface finish and substantial reduction in the cutting forces, the spherical diamond particles (30 nm) at 1, 2, and 4 vol. % in paraffin oil and vegetable oil were used to create the ball-bearing effect in mesoscale grinding of steel SK-41 C  and microscale drilling of aluminum 6061 . In the drilling experiments, the 4 vol. % of nanodiamond particles proved to be more effective with the vegetable oil mixture, while the 1 vol. % of nanodiamond particles was more effective with the paraffin oil mixture. Nevertheless, in the grinding process, the particle size and concentration did not have much influence. Singh et al.  studied the MQL-milling of aluminum AA6061-T6 with ECOCUT SSN 322 neat lubricant oil by adding 0.2 wt. % of SiO2 (5–15 nm). The cutting forces are slightly reduced while the specific energy was substantially reduced up to 40% with the nanoparticle-enhanced lubricant compared to ordinary lubricant oil. Mao et al.  dispersed 0.75 wt. % of Al2O3 particles with the diameter of 60 nm to the de-ionized water in MQL grinding of AISI 52100 steel. In terms of grinding force and surface roughness, the performance of MQL was dependent on the spray parameters such as pressure, spray angle, and nozzle distance. MQL grinding performed as well as flood grinding under the optimum MQL spray parameters .
MQL was applied to the difficult-to-machine materials such as the titanium alloys [25,51–53] and nickel alloys [51,54] with some promising results. Vasu et al.  proved that the Al2O3 nanofluid at 4 and 6 vol. % suspended in a vegetable oil is effective in reducing the cutting temperature in turning of Inconel 600 alloy. The vegetable oil mixed with 6 vol. % of Al2O3 nanoparticles was better in reducing the cutting temperature and improving the surface finish. Kamata and Obikawa  conducted MQL turning of Inconel 718 with synthetic ester mixed with compressed air and argon gas. The nose wear with MQL was less than those in a dry condition and comparable with those of a wet condition. MQL exhibited much better tool life and surface finish (Ra) in comparison with dry and wet lubrications. They also claimed that MQL with compressed air due to the superior properties of heat capacity, thermal conductivity, and lubrication showed better tool life than MQL with argon gas, which was surprisingly worse than the dry machining. Sun et al.  reported the significant improvement in tool life using MQL with EcoCool S-CO5 oil to both air and flood cooling in milling Ti–6Al–4V. They claimed that the formation of TiC layer, commonly believed as a diffusion barrier of carbide tool, occurred in dry and mist cutting conditions, not in a flood cooling condition. Others [55,56] have used only solid lubricants to enhance their machining processes.
These works suggest that a wide variety of conditions exist and an optimal condition is specific to the solid lubricant used and the machining application involved. The friction and wear behaviors of the MQL lubricant mixed with the various solid lubricants, graphite, and hBN, differing in size and aspect ratio, have been examined by conducting the tribometer test and MQL ball milling experiments. One important parameter not studied in the literature is the effect of the surface characteristics of the cutting tools, which will host these MQL mixtures.
Surface Characterization of Two TiAlN-Coated Inserts.
To study the effect of the surface quality of inserts on the performance of the enhanced lubricants, two different TiAlN coated end ball inserts denoted A and B with the exactly same geometry (ZPFG250-PCA12M) were selected. The topography of the tool surfaces was studied using the SEM and profilometer for each tool. Figures 2(a) and 2(b) presents the surface micrographs at 1000×magnification of tools A and B. The roughness parameter was also measured with the Dektak 6M stylus surface profilometer with the point-to-point resolution of 1 Å. These data were attained with the horizontal resolution of a 0.067 μm/sample over the average value of 30,000 data points on the 2 mm evaluation length scan in the parallel and perpendicular directions of the grinding marks on the tool inserts. Figure 2(c) presents the ten-point height mean roughness (Rz) of tool A and tool B measured in both parallel and perpendicular to the grinding marks. Clearly, tool A is considered to have smoother surfaces, which would expect to reduce wear.
Nano- and Micro-platelets Characterizations.
As stated, this paper distinguishes each solid lubricant as either particles or platelets based on its respective aspect ratio, and it defines “nanoplatelets” with the thickness less than 0.1 μm and “microplatelets” with the thickness more than 0.1 μm. The solid lubricants tested in this work include one nanographite particle, two microplatelets (graphite and hBN), and four nanoplatelets (xGnPs and exfoliated hBN). Table 2 summarizes the diameter and thickness of each platelet or particle with their respective supplier. The nanographite particles were procured from ACS Material, LLC, (Medford, MA). The graphite microplatelets were obtained from Alfa Aesar Com. (Ward Hill, MA), while the hBN microplatelets were obtained from Changsung Corp. (Seoul, South Korea). Four nanoplatelets designated as xGnP M5, xGnP C300, xGnP C750, and one exfoliated hBN300 were provided by XG Science, Inc., (Lansing, MI) using their cost-effective exfoliation process by continuously reducing the thickness into a nanoscale by the pulverization process . The hBN300 was made specifically for this work by Dr. I. Do at XG Science, Inc. Among the nanoplatelets, C300 and C750 have the same diameter with distinct thicknesses. Some of these data in Table 2 were interpreted based on the surface area measurements or the SEM images. Figure 3 presents the micrographs of the selected micro- and nano-platelets with the magnification of 1000×. The thickness of nanoplatelets and microplatelets were compared readily in Fig. 4 at the magnification of 30,000×. The thickness of xGnP and hBN300 are much smaller than those of microplatelets. The main advantage of the nanoplatelets is their ability to land, due to their high aspect ratio, on the tool and workpiece surfaces strategically to enhance the sliding between tool and work material after the mixture is sprayed. In addition, the nanoplatelets are expected to cover more area, thus, facilitating a better lubricity and reducing wear even at the same concentration level mixed in the oil.
Enhanced MQL Mixtures.
The vegetable-based oil used in this study is UnistCoolube® 2210 with the flash point around 200 °C from Unist, Inc., (Grand Rapid, MI). It is derived from the rapeseed-based oil. As a lubricant, it reduces the friction between the tool and work material, thus reducing the heat generation. Due to the low flash point, the MQL application of the oil is typically limited in the lower temperature range in interrupted machining such as milling. The enhanced lubricants for MQL were prepared by mixing the vegetable oil with nanoparticles of graphite, micro- and nano-platelets of graphite, and hBN in various contents (up to 5 wt. %) using a high-speed mixer DAC 150FVZ-K from FlackTek (Landrum, SC).
Suspension Stability of Wetting Performance of Mixtures.
In practice, the stability of the mixtures of oil with both particles and platelets of solid lubricant is very important. The performance of enhanced lubricant can be improved only if the mixture is stable throughout the MQL process. The stability is influenced by the shape, dimension (diameter and thickness), and content of the additives and the mixing conditions such as speed and time. The stability test was conducted on the mixtures made with nanoparticles of graphite, micro- and nano-platelets of graphite, and hBN at 0.1 wt. % in the oil. The mixing condition was set at the speed of 2300 rpm for the duration of 8 min after which the stability of the mixtures was achieved. The wettability of the enhanced MQL mixtures may be a crucial factor in the performance in the MQL machining process. The wettability was measured by the wetting angle. The lubricant with a small wetting angle spreads out on the machined surface better to cover a larger area by a thin film between the tool and workpiece. Park et al.  reported the improvement of the wetting angle when the lubricant is mixed with the nanoplatelets. The lubricant with a small wetting angle is expected to stay on a rotating cutting tool better. To examine the differences in wettability among various MQL mixtures, the wetting angles were measured by dropping 0.5 μl of the mixtures on the surfaces of the TiAlN-coated tools that will be used in the ball-milling tests. The wetting angle of a droplet was captured right after applying the droplets on the tool surface by a CCD camera.
Tribological Properties of Mixtures on the Surface of Tool A.
The friction measurement and wear resistance with the enhanced MQL mixtures were attained with the tribological tests. The friction tests were performed with a linear ball-on-disk type tribometer (CSM Instruments) where the 440 c steel ball with 6.35 mm in diameter oscillated on the surface of TiAlN-coated carbide to evaluate the frictional characteristics under various lubrication conditions. All the tests were set at a track length of 6 mm while carrying the loading at 1, 5, and 10 N and the sliding speed at 1.0, 2.5, and 4.0 cm/s at room temperature. The lubrication conditions included dry, pure vegetable oil, and the vegetable oil mixtures with 0.1, 0.5, and 1.0 wt. % of graphite nanoparticles, micro- and nano-platelets of graphite, and hBN. In addition, the wear tests were conducted with a WC ball instead of the steel ball used in the friction tests in order to accelerate the wear on tool surface. The wear characteristics (depth and width of each wear track) were measured after 25,000 and 50,000 cycles. However, because it is not possible to continue the wear test on the same location, the wear cycle started fresh to reach 50,000 cycles. The average depth and width of each wear track were calculated from three cross sections of wear track using a Dektak 6 M stylus surface profilometer. Table 3 summarizes the conditions of our tribological tests. The configuration of the tribological test in relation to the wear track (width (W) and length (L)) is shown in Fig. 5.
MQL Ball Milling Experiments With Tool A.
The ball-mill experiments were conducted on the three-axis vertical Sharnoa CNC Milling Center under dry and enhanced MQL lubrication conditions. A block of AISI 1045 steel (203.2 mm × 127 mm × 203.2 mm) and 25 mm-diameter end ball TiAlN coated carbide inserts (Tool A) were used for the work material and the cutting tool for the experiment as shown in Fig. 6. The cutting process started at one corner of the work material in the direction of 203.2 mm and continued line by line for each pass. A layer is removed when all the passes at the same height are completed. The cutting volume for each layer is 25.8 cm3. The series of MQL ball-mill tests were carried out with the oil mixed with nanoparticles of graphite, nano- and micro-platelets of graphite, and hBN to study the effects of the shape, size, and type of solid lubricants on the wear performance. The lubricants used included pure vegetable oil, the vegetable oil mixed with nanoparticles of graphite (0.1 wt. % of GraP10), microplatelets (0.1 wt. % of graphite, 0.5 wt. % of hBN5) and nanoplatelets (0.1 wt. % of xGnP M5, xGnPC300, and xGnP C750 and 0.5 wt. % of hBN300). The MQL dispensing device (Uni-MAX) provided by Unist, Inc. (Grand Rapid, Michigan) was used to provide the mist in the cutting area. The device sprays the vegetable oil through an external co-axial nozzle. Figure 7 shows our experimental set-up with the pitch and yaw angles of the MQL nozzle. Based on the cutting depth and the location of the nozzle employed throughout the experiment, the pitch angle is needed to be more than 11 deg for the oil-mist to access the cutting zone and thus was fixed at 15 deg throughout the experiment. The yaw angle was set at 180 deg, covering larger flank and central zones by the oil-mist during the MQL cutting process (see Sec. 5.3). The nozzle outlet pressure and flow rate can be adjusted with the air pressure screw and pulse duration/frequency in the control panel, respectively. As we did in the previous work [23,24], the optimum MQL condition was set to be the output pressure of 8 psi and a flow rate of 1.5 ml/min with our MQL dispensing device. The flank wear was also measured after cutting each layer to record the progress of tool wear. The machining conditions are summarized in Table 4.
Results and Discussion
Suspension Stability and Wetting Performance of Mixtures.
The stability of each mixture right after mixing was observed after 1, 6, 24, and 72 h. Due to the space constraint, the mixtures with 0.1 wt. % of nanoparticles, micro- and nano-platelets in 72 h (3 days) after the mixing are shown in Fig. 8. The mixture of the platelets with smaller thickness and diameter was more stable. The tests showed that the microplatelets mixtures were stable only less than 1 day after mixing in comparison to more than 3 days for the nanoplatelets mixtures. The stability tests with graphite and hBN platelets show that the thickness influenced the stability more than the diameter. For example, the microplatelets of graphite with the thickness of 200 nm and the averaged diameter of 7 μm were segregated much faster than the nanoplatelet, xGnP M5, with the thickness of 6–8 nm and the diameter of 5 μm. Our previous work  pointed that the segregation of xGnPs was affected by the diameter of the platelets based on the observation that one xGnP grade was segregated quicker than the other xGnP with a smaller diameter despite the same thickness.
Figure 9 indicates the wetting angles of water, mineral oil with water, mixtures of nanoparticles of graphite, micro- and nano-platelets of graphite, and hBN at a room temperature. The pure vegetable oil showed the smallest wetting angle compared to pure water and mineral oil mixed with water. It can be seen that not only the introduction of nanoparticles graphite, micro- and nano-platelets decreases the wetting angles but also that the wetting angle of mixture with nano-hBN platelets is slightly smaller than those of xGnP mixtures. This indicates that the droplets of nanoparticles, micro- and nano-platelets mixtures can spread out better on the tool surface leading to larger area of the tool was covered by more uniform lubricant film. The adhesion of those droplets also was improved, which helped the droplets to stay on the tool surface at high speed.
Lubrication Properties and Wear Resistance Performance.
The friction coefficients at the sliding speed of 2.5 and 4.0 cm/s under various loads are summarized in Fig. 10. The friction results show that the oil primarily controls the friction behavior. However, at the elevated temperatures expected in MQL machining, solid lubricants are expected to provide a better lubricant performance . With the increase in load, the friction has increased slightly in general. As expected, the friction coefficients of oil mixture were decreased with the increase in speed.
The results of wear test by sliding the WC ball on the TiAlN coated surface of tool A are summarized in Fig. 11. The track depth is much more pronounced than the track width. The wear tracks generated with the nanoplatelets mixture were smaller than those with the nanoparticles and microplatelets mixture in both the depth and width of the wear tracks. On the other hand, the microplatelets did not perform perhaps due to the larger thickness and the reduction in the covered area of the solid lubricants. The nanoparticles of graphite with a particle size of 10 nm slightly improved in wear resistance. The tiny spherical particle of the graphite may roll at the interface between the WC ball and tool A, while the nanoplatelets due to their high aspect ratio “stick” at the interface providing low friction and wear resistance. The result from our wear test reveals the preponderant wear resistance of the nanoplatelets of hBN and xGnP. Among the three grades of xGnP (M5, C300, and C750), C750 with the smallest thickness of 1.2 nm performed the best because the coverage area is expected to be the maximum, which suggests the importance of the thickness. In terms of the weight content, the nanoplatelets are expected not only to effectively cover the larger area of the tool surfaces but also to facilitate the sliding in their planar direction between the WC ball and a tool surface. Because of their thickness, more than 100 nanoplatelets or only one or two microplatelets can be stacked on each “valley” on the tool surface as shown in Fig. 12. In reality, the nanoplatelets cover a larger area, which enables them to provide better lubricity and resistance to wear. Overall, the wear performance agrees well with the ball-on-disk study of Berman et al. . The multilayered graphene suspended in ethanol (1 mg/L) has reduced the wear rate on a 440 C grade steel ball more than 100 times without an ethanol solution and 15 time with only ethanol, although the friction coefficients are reduced by four times without an ethanol solution and 1.6 times with only ethanol. With the application of the nanosolution every 400 cycles, the wear rates are substantially reduced (>1000 times). Kimura et al.  conducted the sliding experiment on bearing steel against itself and cast iron with the hBN-mixed lubricant and presented a slight improvement in friction and a significant improvement in wear.
Flank Wear in MQL Process.
Figure 13 presents the flank wear results with Tool A from the MQL milling experiment. Surprisingly, the microplatelets-enhanced lubricants resulted in more flank wear than pure MQL oil (traditional MQL), exposing the ineffectiveness of the microplatelets, which must contribute to flank wear. The nanoplatelets of graphite showed significant reduction in flank wear compared to pure MQL oil. Comparing three xGnP grades (M5, C300, and C750), both C300 and C750 grades, despite the chipping after cutting five layers (cutting volume of 129.1 cm3), significantly reduced flank wear than M5. Both C300 and C750 significantly improved the wear performance compared to M5. In fact, the flank wear with 0.1 wt. % of xGnP M5 was slightly reduced compared to the pure oil. Similarly, the comparison between hBN5 (microplatelets) and hBN300 (nanoplatelets) also indicates the importance of the nanoscale thickness even though a higher content (0.5 wt. %) was necessary. The hBN300 showed a slightly better wear performance than xGnP C750 due to the higher dissociation temperature of hBN. For tool A with the smooth surfaces, maintaining the thickness of solid lubricants in a nanoscale is critically based on the beneficial effects with xGnP M5, xGnP C300, xGnP C750, and hBN300 as shown in Fig. 13. This similar trend was also observed in the tribological tests shown in Fig. 11.
Verification of the Results
To verify the superior performance of the nanoplatelets-enhanced mixtures, more experiments with the tribotests and MQL milling were conducted with another tool, tool B with the rougher coated surface. In addition, another experimental set was carried out for finding the optimal yaw angle to enhance the performance of the MQL process. To reduce the number of tests, only a few selected micro- and nano-platelets mixtures were selected.
Tribotest of MQL Mixtures With Tool B.
The wear resistance was conducted on the rough surface of tool B with the enhanced MQL mixtures with hBN5 and hBN300. Due to the roughness, the wear track data on tool B show only slight improvement with the nanoplatelets mixtures compared to microplatelets as shown in Figs. 14(a) and 14(b). The higher weight content of both micro- and nano-platelets performed slightly better. In Fig. 14(c), the result of the friction coefficient in the wear tests also showed no difference between micro- and nano-platelets of hBN, while the friction was slightly reduced at the 50,000 cycles as the sliding surface may become smoother during the test.
Tool Wear Performance of MQL Mixtures With Tool B.
For the case of tool B as shown in Fig. 15, three solid lubricant mixtures were compared to pure oil. The MQL-ball mill experiments with both nano- and micro-platelets-enhanced lubricants performed better than pure oil lubricant (traditional MQL) with the cutting edge exhibiting gradual wear. However, the higher content of microplatelets did not yield a better performance. With the xGnP M5 nanoplatelets, Park et al.  also found that the mixture with 0.1 wt. % performed better than the mixture with 1.0 wt. % when machining with tool A. Because of the rougher surface of tool B, a higher concentration (1.0 wt. %) of the xGnP M5 was better as our result indicated in Fig. 15. Surprisingly, the microplatelets such as hBN5 are beneficial on tool B, especially the mixture with 0.5 wt. % of hBN5. The thicker platelets may be needed to make up for the deeper valley shown in Fig. 12. This is not the case for the smooth surface of tool A as the hBN5 mixture was detrimental as shown in Fig. 13.
Improvement of Enhanced MQL Mixtures With the Optimal Spray Yaw Angle.
As discussed earlier, the performance of the MQL process can be improved using the mixtures with only nanoplatelets of graphite and hBN. In another aspect, the performance of MQL depends directly not only on the MQL mixture but also on the interaction with the tool during the cutting process. The interaction is depended mainly on the spray angle for a given tool geometry. To determine the optimal yaw angle, the MQL ball milling tests were conducted with tool B at nozzle yaw angles of −30, −90, −150, 60, 120, and 180 deg to the X-axis. The resulting flank wear is plotted at various yaw angles in Fig. 16. The best result was attained with the yaw angles of 180 deg and −30 deg. For the −30 deg yaw angle setup, the rake face of tool and cutting zone were exposed to the MQL spray during process. With the 180 deg yaw angle, even though the cutting zone is blocked by the tool, the flank face of tool is always exposed to the MQL spray throughout the process. This finding is corroborated by the geometry calculation, which indicates a larger flank and central zone exposed to the MQL spray during the cutting process. Figure 17(a) shows that the negative 30 deg yaw angle setup provided a relatively good lubrication condition, while Fig. 17(b) shows that the 120 deg yaw angle configuration showed the worst condition. Based on this finding, the pitch angle of 15 deg and the yaw angle of 180 deg were selected for the MQL ball-mill tests.
The results of the tribotests and ball milling experiments with platelets-enhanced MQL lubricants lead to following conclusions:
In terms of friction, the solid lubricant additives in vegetable oil did not show any positive effect in reducing the friction at room temperature regardless of the shape size and concentrations of additives. The base oil controls the friction mainly at room temperature.
In the sliding wear test, the nanoparticle and microplatelets-mixed lubricants did not present any advantage over the pure oil. The nanoplatelets-enhanced lubricants performed better due to the lubricity provided by the platelets at the interface between two surfaces. However, no significant wear reduction was observed with these nanoplatelets-enhanced lubricants on the rough surface.
The MQL enhanced by solid lubricant presented significant improvement in wear performance with the nanoplatelets (xGnP M5, xGnP C300, xGnP C750, and hBN300) on both rough or flat tool surfaces. The microplatelets (Graphite and hBN5) performed worse than pure oil (traditional MQL process) and the nanoparticles (GraP10) show slight improvement compared to the pure oil. However, with the rough tool surfaces, the MQL lubricant mixed with the small amount of microplatelets (hBN5) has the beneficial effect. In addition, a much higher concentration is required to enhance the performance for the rough surface as the flank wear was reduced by about 30% with tool B with 1.0 wt. % of xGnP M5.
The orientation of the MQL nozzle tremendously affects the MQL process. We determined that the optimal yaw spray angle could be −30 deg or 180 deg which provided the cutting zone well-lubricated.
It is also important to recognize that finding an optimum solid lubricant to enhance MQL oil depends on not only the characteristics (material as well as morphology) of solid lubricants but also on the characteristics of the tool surface. In addition, only a small amount (e.g., 0.1 wt. % for tool A) of nanoplatelets, which make this process a cost-effective solution with the existing equipment. More importantly, a wide variety of shape, thickness, and size/size distribution of the solid lubricants including nanoplatelets are now available and many reports on new processing techniques, which will eventually make the solid lubricants cheaper and available for wide varieties of applications.
This work was supported by Korea Institute of Industrial Technology (KITECH, JA-15-0034) and the Ministry of Knowledge Economy (MKE) in Republic of Korea (Project title is “Development of liquid nitrogen-based cryogenic machining technology and system for titanium and CGI machining”, 10048871). Some of the equipment was available through Fraunhofer CCL, (East Lansing, MI) and we also want to acknowledge Dr. I. Do at XG Science, Inc. (Lansing, MI) for providing the nanoplatelets, hBN 300, specifically for this work.