materials
Article A Study on Machining Performances of Micro-Drilling of Multi-Directional Carbon Fiber Reinforced Plastic (MD-CFRP) Based on Nano-Solid Dry Lubrication Using Graphene NanoPlatelets
Jin Woo Kim 1, Jungsoo Nam 2, Jaehun Jeon 1 and Sang Won Lee 3,*
1 Department of Mechanical Engineering, Graduate School, Sungkyunkwan University, Suwon-si 16419, Korea; [email protected] (J.W.K.); [email protected] (J.J.) 2 Intelligent Manufacturing System R&D Department, Korea Institute of Industrial Technology, Cheonan-si 31056, Korea; [email protected] 3 School of Mechanical Engineering, Sungkyunkwan University, Suwon-si 16419, Korea * Correspondence: [email protected]; Tel.: +82-31-290-7467
Abstract: The objective of this study is to investigate the tribological behavior of graphene nanoplatelets (xGnPs) as nano-solid lubricants, and to evaluate their applicability to the micro-drilling of multi- directional carbon fiber-reinforced plastic (MD-CFRP). To verify the tribological effect of nano-solid lubricants, three kinds of xGnPs (xGnP C-750, xGnP M-5, and xGnP H-5), multiwall carbon nanotubes (MWCNTs), and hBN are compared by the ball-on-plate test. Of these, three xGnPs are selected as nano-solid lubricants to investigate the micro-drilling performance of MD-CFRP using nano-solid dry
lubrication, and the experimental results demonstrate that all xGnPs can enhance lubrication action in terms of surface quality (delamination, uncut fiber, and inner surface) and tool wear. In particular, larger graphene nanoplatelets (xGnP M-5 and xGnP H-5) are superior to the smaller one (xGnP C-750) Citation: Kim, J.W.; Nam, J.; Jeon, J.; Lee, S.W. A Study on Machining by guaranteeing enhanced sliding action between the tool grain and the CFRP composite. Performances of Micro-Drilling of Multi-Directional Carbon Fiber Keywords: micro-drilling; multi-directional carbon fiber-reinforced plastic (MD-CFRP); nano-solid Reinforced Plastic (MD-CFRP) Based dry lubrication; tribological behavior; machining performance; graphene nanoplatelets on Nano-Solid Dry Lubrication Using Graphene NanoPlatelets. Materials 2021, 14, 685. https://doi.org/ 10.3390/ma14030685 1. Introduction Generally, a carbon fiber-reinforced plastic (CFRP) composite that can be fabricated Received: 5 January 2021 by combining carbon fiber and polymer matrix has been used for high-tech applications, Accepted: 28 January 2021 such as in the automobile and aerospace industries. The CFRP composite is generally Published: 2 February 2021 known as a superior metal substitute due to its advanced properties, such as superior strength-to-weight ratio, stiffness, damage tolerance, and so on [1,2]. Publisher’s Note: MDPI stays neutral However, because of its complicated structure, which is represented by anisotropy with regard to jurisdictional claims in and non-homogeneity, the use of CFRP in machining processes can lead to severe surface published maps and institutional affil- iations. defects such as delamination, fiber pull-out [3,4]. In addition, due to frictional force and high temperature at the tool-work interface, excessive tool wear occurs [5]. Therefore, to enable efficient machining with CFRP composite, there is a demand for an appropriate lubrication approach. When machining metallic materials, metal working fluids (MWFs) has typically been Copyright: © 2021 by the authors. used as the types of wet flooding and minimum quantity lubrication (MQL) for reducing Licensee MDPI, Basel, Switzerland. high cutting heat, and improving lubrication performance, respectively. To further enhance This article is an open access article machining performance, nano-scale particles have been used as nanofluids by adding distributed under the terms and conditions of the Creative Commons them to MWFs and liquid medium to help the lubrication action during the machining Attribution (CC BY) license (https:// process, and their effectiveness in nanofluid MQL machining on metallic materials has creativecommons.org/licenses/by/ been verified [6–8]. 4.0/).
Materials 2021, 14, 685. https://doi.org/10.3390/ma14030685 https://www.mdpi.com/journal/materials Materials 2021, 14, 685 2 of 19
However, it has been reported that CFRP composite has moisture-sensitive properties, and long-term exposure to liquid could negatively affect its excellent mechanical proper- ties [9–11]. Moreover, there is a lack of research on appropriate lubrication technologies that could prevent degradation of CFRP’s mechanical properties, and it was also found that the wet machining might worsen tool life and hole quality during the CFRP drilling [12,13]. Considering the moisture-sensitive nature of CFRP, the author’s research group has recently proposed a novel lubrication technique that has been named the nano-solid dry lubrication technique [14]. With this new method, air-driven nano-scale particles can be sprayed into the cutting zone. In this study, one kind of graphene nanoplatelets (xGnP H-5) and multiwall carbon nanotubes (MWCNTs) were used as nano-solid lubricants, respectively, and their machining performances on uni-directional CFRP (UD-CFRP) were analyzed in terms of surface quality and tool wear. The results showed that xGnP H-5 was more advantageous than MWCNT. In a previous study, an analysis considering the tribological behavior of nano-solid lubricants was not conducted. To elucidate the lubrication mechanism of carbon allotrope particles, it is necessary to analyze the tribological properties of the nanoparticles. In various studies involving tribological tests, the types of using carbonous lubricants used to lubricate the two-body interface can be classified into four groups (I. specimen coating, II. specimen reinforcement as additive, III. nanofluid, and IV. solid powder).
2. Tribological Test of Nano-Solid Lubricants Prior to the micro-drilling experiments, to confirm the effects of the graphene nano- platelets as lubricants against CFRP composites, a series of triobological tests was con- ducted via the ball on plate test, and the coefficient of friction and magnitude of wear were analyzed. In addition to the graphene nanoplatelets, multiwall carbon nanotubes (MWCNTs) and hexagonal boron nitride (hBN) were also investigated for a comparison. As a result, five nano-solid lubricants were used in total for the tribological test.
2.1. Properties of Nano-Solid Lubricants The structures and shapes of the nanoparticles were confirmed by using field emis- Materials 2021, 14, 685 sion scanning electron microscope (JSM-7600F, JEOL Ltd., Tokyo, Japan), and the images3 of 19 are shown in Figure 1. In this paper, while taking field emission scanning electron micro- scope (FE-SEM) images, the acceleration voltage was 15.0 kV. In addition, the platinum (Pt)2. Tribological sputtering was Test conducted of Nano-Solid for minimizing Lubricants a charging effect. The properties of the five nano-solid lubricants are summarized in Table 1. The three Prior to the micro-drilling experiments, to confirm the effects of the graphene nanoplatelets different graphene nanoplatelets called xGnP manufactured by XG SCIENCE were clas- as lubricants against CFRP composites, a series of triobological tests was conducted via sified as xGnP C-750, xGnP M-5, and xGnP H-5. The xGnP C-750 was categorized as C the ball on plate test, and the coefficient of friction and magnitude of wear were analyzed. grade with an average particle size (APS) of ~0.2 μm and a specific surface area (SSA) of In addition to the graphene nanoplatelets, multiwall carbon nanotubes (MWCNTs) and 750 m2/g. The xGnP M-5 was larger than the xGnP C-750 and had an APS of 5 μm as well hexagonal boron nitride (hBN) were also investigated for a comparison. As a result, five as a thickness of 8 nm. The xGnP H-5 was the largest of the three graphene nanoplatelets, nano-solid lubricants were used in total for the tribological test. and it had an APS of 5 μm and a thickness of 15 nm. It was confirmed that the three xGnP particles2.1. Properties had a ofform Nano-Solid of stacks Lubricants of thin two-dimensional (2D) sheets in common. In addition, MWCNTs (US4500) manufactured by US Research Nanomaterials, Inc. The structures and shapes of the nanoparticles were confirmed by using field emission (Houston,scanning TX, electron USA) microscope which consist (JSM-7600F, of carbon JEOL(C) allotropes Ltd., Tokyo, such Japan),as xGnPs and but the with images a tube- are shapeshown structure in Figure were1. In confirme this paper,d, and while the takinghBN (MK-hBN-070) field emission from scanning M K Impex electron Corp. microscope (Missis- sauga,(FE-SEM) Canada), images, which the is acceleration simply known voltage as white was graphene 15.0 kV. Inwith addition, an analogous the platinum structure (Pt) to xGnPsputtering by bonding was conducted boron (B) and for minimizingnitrogen (N), a are charging presented effect. and summarized in Figure 1 and Table 1.
(a) xGnP C-750 (b) xGnP M-5 (c) xGnP H-5
(d) MWCNTs (e) hBN
FigureFigure 1. 1. FE-SEMFE-SEM images images of of xGnP xGnPs,s, MWCNTs, MWCNTs, and and hBN. hBN.
The properties of the five nano-solid lubricants are summarized in Table1. The three different graphene nanoplatelets called xGnP manufactured by XG SCIENCE were classified as xGnP C-750, xGnP M-5, and xGnP H-5. The xGnP C-750 was categorized as C grade with an average particle size (APS) of ~0.2 µm and a specific surface area (SSA) of 750 m2/g. The xGnP M-5 was larger than the xGnP C-750 and had an APS of 5 µm as well as a thickness of 8 nm. The xGnP H-5 was the largest of the three graphene nanoplatelets, and it had an APS of 5 µm and a thickness of 15 nm. It was confirmed that the three xGnP particles had a form of stacks of thin two-dimensional (2D) sheets in common. In addition, MWCNTs (US4500) manufactured by US Research Nanomaterials, Inc. (Houston, TX, USA) which consist of carbon (C) allotropes such as xGnPs but with a tube-shape structure were confirmed, and the hBN (MK-hBN-070) from M K Impex Corp. (Mississauga, Canada), which is simply known as white graphene with an analogous structure to xGnP by bonding boron (B) and nitrogen (N), are presented and summarized in Figure1 and Table1. Materials 2021, 14, x FOR PEER REVIEW 4 of 20
Materials 2021, 14, 685 4 of 19 Table 1. Properties and shapes of xGnPs, MWCNTs, and hBN [19–21].
xGnP C-750 xGnP M-5 xGnP H-5 MWCNTs hBN Table 1. Properties and shapes of xGnPs, MWCNTs, and hBN [19–21]. Outer diameter: 10~30 APS: ~0.2 μm APS: 5 μm APS: 5 μm APS: 70 nm xGnP C-750 xGnP M-5 xGnP H-5 MWCNTsnm hBN Inner diameter: 5~10 Thickness:APS: ~0.2 ~1 µnmm APS: Thickness: 5 µm ~8 nm APS: 5 µm Thickness:~15 Outer nm diameter: 10~30 nm APS:SSA: 70 19 nm m2/g nm Thickness: ~1 nm Thickness: ~8 nm Thickness: ~15 nm Inner diameter: 5~10 nm SSA: 19 m2/g SSA: 750 m2/g SSA: 120~150 m2/g SSA: 50~80 m2/g Length: 10~30 μm SSA: 750 m2/g SSA: 120~150 m2/g SSA: 50~80 m2/g Length: 10~30 µm 2 SSA: >200 SSA: m2/g >200 m /g Shape: Lamella struc- Shape:Shape: 2D-Sheet 2D-Sheet Shape: Shape: 2D-Sheet 2D-Sheet Shape: 2D-Sheet Shape: 2D-Sheet Shape: Tube-shape Shape: Tube-shape Shape: Lamella structure ture Element: Boron (B), Element: Carbon I Element: Carbon I Element: Carbon I Element: Carbon I Element: Boron (B), Element: Carbon (C) Element: Carbon (C) Element: Carbon (C) Element: Carbon (C) Nitrogen (N) Nitrogen (N)
InIn order order to to confirm confirm the the degree degree of of crystal crystallinitylinity and and the the number number of of graphene graphene layers, layers, RamanRaman spectroscopy spectroscopy equipment (Micro-Raman(Micro-Raman SpectroscopySpectroscopy System, System, RENISHAW, RENISHAW, Wotton- Wot- ton-under-Edge,under-Edge, England) England) was was used. used. In In Figure Figure2, 2, Raman Raman spectra spectra of of xGnPs xGnPs andand MWNCTsMWNCTs areare given. given. Generally, Generally, the the D Dband band (1300–1400 (1300–1400 cm cm−1) −is1 )related is related to the to edge the edge disorder disorder and andde- fectsdefects in carbon in carbon structure, structure, and andthe G the band G band (1550~1615 (1550~1615 cm−1) cmis known−1) is known as the graphitic as the graphitic region. Theregion. 2D band The 2D (~2700 band cm (~2700−1) is cman −overtone1) is an overtone of the D of band, the D and band, its andasymmetric its asymmetric shape is shape the indexis the of index multi-layer of multi-layer of graphene. of graphene. The ratio The between ratio between intensity intensity of the D of band the D and band the and G bandthe G (I bandD/IG) is (I Dused/IG )to is an used indicator to an indicator of crystallinity of crystallinity of the graphene of the graphene layer. The layer. ratio Thebetween ratio intensitybetween of intensity the 2D ofband the and 2Dband G band and (I G2D/I bandG) explains (I2D/IG the) explains number the of numbergraphene of layers. graphene In Figurelayers. 2, In it Figurewas found2, it was that found MWCTNs that MWCTNs in the form in of the tube-shape form of tube-shape had larger had ID/I largerG than IthoseD/IG ofthan xGnPs, those indicating of xGnPs, a indicatingdecrease in a the decrease degree in of the crystallinity degree of of crystallinity the graphene of thelayer. graphene When comparinglayer. When the comparing I2D/IG of xGnPs, the I2D /ItheG ofvalues xGnPs, less the than values 1 was less confirmed, than 1 was which confirmed, means which the multi-layermeans the multi-layergraphene (graphite). graphene In (graphite). addition, In the addition, asymmetry the asymmetry of the 2D bands of the 2Dwas bands con- firmedwas confirmed in the large in thexGnPs large (xGnP xGnPs M-5 (xGnP and xG M-5nP and H-5) xGnP compared H-5) comparedto smaller toone smaller (xGnP oneC- 750)(xGnP [22–25]. C-750) [22–25].
FigureFigure 2. 2. RamanRaman spectra spectra of of xGnPs xGnPs and and MWCNTs MWCNTs. 2.2. Tribological Test Design and Conditions 2.2. Tribological Test Design and Conditions In order to investigate the tribological properties of nano-solid lubricants against In order to investigate the tribological properties of nano-solid lubricants against MD-CFRP composites, reciprocating ball on plate equipment (TE77 AUTO, Phoenix Tribol- MD-CFRPogy Ltd., Kingsclere,composites, England) reciprocating was used,ball on and plate the equipment coefficient of(TE77 friction AUTO, (COF) Phoenix datawere Tri- bologymeasured Ltd., in Kingsclere, each test case. England) Figure was3 shows used, the and ball the on coefficient plate equipment of friction containing (COF) data the were test measuredspecimen in and each the test test case. ball. Figure 3 shows the ball on plate equipment containing the test specimenThe testand specimenthe test ball. was fabricated with MD-CFRP, which is prepared by laminating prepreg with a carbon fiber of T700 (Toray Composite Materials America, Inc., Tacoma, WA, USA) with 13-ply, and the size was 38 mm × 58 mm × 4 mm. The stacking sequence ◦ ◦ ◦ ◦ ◦ ◦ ◦ ◦ ◦ ◦ ◦ ◦ ◦ of the MD-CFRP specimen was [0 /0 /90 /90 /90 /0 /0 /0 /90 /90 /90 /0 /0 ]s, and information on the carbon fiber and the prepreg is listed in Table2. The test ball was tool steel with a diameter of 10 mm, and a new test specimen and ball were used for each experiment.
Materials 2021, 14, 685 5 of 19 Materials 2021, 14, x FOR PEER REVIEW 5 of 20
Figure 3. Ball on plate test equipment.equipment
Table 2. Information on carbon fiber and prepreg [26]. The test specimen was fabricated with MD-CFRP, which is prepared by laminating prepregProperties with a of carbon Carbon fiber Fiber of and T700 Prepreg (Toray Composite Materials Information America, Inc., Tacoma, WA, USA) with 13-ply, and the size was 38 mm × 58 mm × 4 mm. The stacking sequence Thickness of each ply 310 µm of the MD-CFRPStandard specimen of carbon was fiber [0°/0°/90°/90°/90°/0°/0°/0°/90°/90°/90°/0°/0°] T700 s, and infor- mation on the carbonTensile strengthfiber and the prepreg is listed in Table 2. 4900The Mpatest ball was tool steel with a diameterTensile of 10 modulusmm, and a new test specimen and ball were 230 Gpa used for each experi- ment. Strain 2.1% Density 1.80 g/cm3 Table 2. InformationFilament on carbon diameter fiber and prepreg [26]. 7 µm Fiber area weight (FAW) 150 g/m2 Properties of CarbonFiber volume Fiber and Prepreg Information 64% Thickness of each ply 310 μm In total,Standard six experimental of carbon casesfiber were tested, consisting of a dry T700 case without lubricants, xGnP C-750, xGnPTensile M-5, strength xGnP H-5, MWCNTs, and hBN. Except 4900 for MPa the dry case, 0.05 g of lubricants wasTensile sprinkled modulus on each MD-CFRP specimen. The load 230 of GPa the test ball (normal force) was 50 N, andStrain the sliding distance was 15 mm. In addition,2.1% the reciprocating speed and the total test timeDensity were 2 Hz, and 610 s, respectively. 1.80 g/cm3 Filament diameter 7 μm 2.3. Tribological Test Results Fiber area weight (FAW) 150 g/m2 In order toFiber analyze volume the tribological behaviors of nano-solid 64% lubricants on the CFRP composite, the coefficient of friction (COF) and the wear on the specimen were investigated. In eachIn total, case, six the experimental COF value was cases averaged were test fored, 180 consisting sec after theof a testdry was case stabilized. without lubri- The averagecants, xGnP COF C-750, values xGnP according M-5, to xGnP different H-5, nano-solid MWCNTs, lubricants and hBN. are Except given infor Figure the dry4a, case, and these0.05 g areof lubricants compared was with sprinkled that of the on dry each case. MD-CFRP As shown specimen. in the graph, The load the of highest the test value ball was(normal obtained force) in was the 50 dry N,case and withoutthe sliding lubricants. distance Inwas the 15 cases mm. ofIn theaddition, three xGnPsthe reciprocat- and the MWCNTs,ing speed and the the values total were test significantlytime were 2 Hz, reduced. and 610 On s, the respectively. other hand, the COF in the case of hBN showed little difference compared to the dry case. In Figure4b, the average wear width2.3. Tribological and depth Test values Results are given. Similar to the COF results, the wear width and depth of the specimenIn order to was analyze substantially the tribological smaller behaviors in the cases of nano-solid of xGnPs andlubricants MWCNTs on the composed CFRP com- of graphene-basedposite, the coefficient particles. of friction On the (COF) other and hand, the theywear were on the both specimen highest were in the investigated. case of hBN. In each Thesecase, the wear COF values value were was measuredaveraged byfor obtaining180 sec after images the test of the was specimen stabilized. and The 3D average profile dataCOF byvalues using according a 3D laser to different microscope nano-solid (VK-X lubricants series, Keyence, are given Osaka, in Figure Japan). 4a, and The these optical are microscopiccompared with photos that of of the each dry test case. case As areshown shown in the in graph, Figure 5the. For highest more value detailed was analysis, obtained 5 sectors ([a]–[e]) were selected, and they were 150 µm apart vertically from each other. The in the dry case without lubricants. In the cases of the three xGnPs and the MWCNTs, the direction of measurement of wear width was expressed to yellow color, and the maximum values were significantly reduced. On the other hand, the COF in the case of hBN showed wear depth was expressed to red color, as representatively shown in Figure5a. In each case, little difference compared to the dry case. In Figure 4b, the average wear width and depth the wear width and depth were measured and averaged using the VK Analyzer software. values are given. Similar to the COF results, the wear width and depth of the specimen
Materials 2021, 14, x FOR PEER REVIEW 6 of 20
was substantially smaller in the cases of xGnPs and MWCNTs composed of graphene- based particles. On the other hand, they were both highest in the case of hBN.
Materials 2021, 14, x FOR PEER REVIEW 6 of 20
Materials 2021, 14, 685 6 of 19 was substantially smaller in the cases of xGnPs and MWCNTs composed of graphene- based particles. On the other hand, they were both highest in the case of hBN.
(a) Average coefficient of friction (b) Average wear width and wear depth Figure 4. Average coefficient of friction and wear values in each case.
These wear values were measured by obtaining images of the specimen and 3D pro- file data by using a 3D laser microscope (VK-X series, Keyence, Osaka, Japan). The optical microscopic photos of each test case are shown in Figure 5. For more detailed analysis, 5 sectors ([a]–[e]) were selected, and they were 150 μm apart vertically from each other. The direction of measurement of wear width was expressed to yellow color, and the maximum
wear(a) depthAverage was coefficient expressed of to friction red color, as re (presentativelyb) Average wear shown width in and Figure wear 5a. depth In each case, the wear width and depth were measured and averaged using the VK Analyzer soft- Figure 4. Average coefficient of friction and wear values in each case. ware.Figure 4. Average coefficient of friction and wear values in each case. These wear values were measured by obtaining images of the specimen and 3D pro- file data by using a 3D laser microscope (VK-X series, Keyence, Osaka, Japan). The optical microscopic photos of each test case are shown in Figure 5. For more detailed analysis, 5 sectors ([a]–[e]) were selected, and they were 150 μm apart vertically from each other. The direction of measurement of wear width was expressed to yellow color, and the maximum wear depth (wasa) Dry expressed to red color, (b) asxGnP representatively C-750 shown in (c) FigurexGnP M-55a. In each case, the wear width and depth were measured and averaged using the VK Analyzer soft- ware.
(d) xGnP H-5 (e) MWCNTs (f) hBN
FigureFigure 5. 5. OpticalOptical microscopic microscopic images images of of MD-CFRP MD-CFRP after after the the ball ball on on plate plate test. test. (a) Dry (b) xGnP C-750 (c) xGnP M-5 ForFor more more detailed analysis,analysis, FE-SEMFE-SEM images images of of the the worn worn surfaces surfaces were were taken, taken, and theirand theirmorphological morphological characteristics characteristics were were studied. studied. In Figure In Figure6, the 6, FE-SEM the FE-SEM images images for each for each case caseare shown,are shown, and and the surfacethe surface morphologies morphologies in the in casesthe cases of three of three xGnPs xGnPs and MWCNTsand MWCNTs were werequite quite similar similar one another.one another. It seems It seems that the that entire the entire surfaces surfaces were coveredwere covered with continuous with con- tinuousfilm. Therefore, film. Therefore, it is assumed it is assumed that the that lubrication the lubrication mechanism mechanism of these of these graphene-based graphene- particles could be similar. On the other hand, in the cases of dry and hBN, such continuous based particles(d) xGnP could H-5 be similar. On (thee) MWCNTs other hand, in the cases of dry(f) hBN and hBN, such Materials 2021, 14, x FOR PEER REVIEW 7 of 20 continuousfilm is not observed,film is not andobserved, the worn and surfaces the worn are surfaces rougher are and rougher has cracks. and has In Sectioncracks. 2.4In , Figure 5. Optical microscopic images of MD-CFRP after the ball on plate test. Sectionfurther 2.4, scientific further explanations scientific explanations are given. are given. For more detailed analysis, FE-SEM images of the worn surfaces were taken, and their morphological characteristics were studied. In Figure 6, the FE-SEM images for each case are shown, and the surface morphologies in the cases of three xGnPs and MWCNTs were quite similar one another. It seems that the entire surfaces were covered with con- tinuous film. Therefore, it is assumed that the lubrication mechanism of these graphene- based particles could be similar. On the other hand, in the cases of dry and hBN, such continuous film is not observed, and the worn surfaces are rougher and has cracks. In (a) Dry (b) xGnP C-750 (c) xGnP M-5 Section 2.4, further scientific explanations are given.
(d) xGnP H-5 (e) MWCNTs (f) hBN
FigureFigure 6. 6. FE-SEMFE-SEM images images of of worn worn surfaces surfaces du duringring the the ball ball on on plate plate test test (× (× 300)300).
2.4. Discussions In these tribological test results, the three graphene nanoplatelets (xGnP C-750, xGnP M-5, and xGnP H-5) and the multiwall carbon nanotubes (MWCNTs) could meaningfully reduce the coefficient of friction and the wear width and depth. However, hexagonal bo- ron nitride (hBN) was not effective. To elucidate the primary lubrication mechanism for each case reasonably, the FE-SEM images given in Figure 6 were more deeply investigated by referencing some studies on tribological characteristics of carbonous materials. The an- alyzed schematic expressions are provided in Figure 7. It has been known that carbonous materials basically have the self-lubricating prop- erty. Suresha et al. reported a self-lubricating nature of carbon fiber from the abrasion test [27]. Zhu et al. conducted the pin joint test of graphite and carbon fiber modified polymer, and the self-lubricating carbonous characteristics were found by observing worn surfaces and wear debris [28]. In addition, these carbonous materials could create tribofilm. Zhao et al. found that short carbon fibers improve the anti-wear properties of polymer compo- sites during the block-on-ring test by forming tribofilm [29]. Luo et al. found the formation of transfer film of polymer-carbon fiber composite by acting dry-sliding during the block- on-ring test [30]. Therefore, in the dry case, wear debris occurred by the breakage of the carbon fiber which composes the MD-CFRP specimen could stay between the two counterparts. These carbonous materials could have formed carbon-based tribofilm due to the repeated stress- ing cycles. As a result, dry-sliding behavior occurred due to the self-lubricating capability of carbonous materials in MD-CFRP, despite the lack of nano-solid lubricants. In this context, the carbonous nano-solid lubricants could enhance such lubricating capability even more. In the cases of the three xGnPs, it was analyzed that the very thin two-dimensional graphene layers (general thickness of single-layer graphene: ~0.34 nm) slid on each other by enabling proper lubrication action. The lubrication characteristics of xGnP C-750, xGnP M-5, and xGnP H-5 were difficult to distinguish from each other, as shown in the test results. This could have been because the particles were rubbed by the continuous movement of reciprocating sliding while loading the normal force on the test ball, so analogous graphene-based tribofilm could be formed in all xGnP cases by increas- ing the contact pressure between the test ball and the specimen. In the case of MWCNTs, the tube-shape structure might have been maintained at the beginning of the lubrication phase, despite the load on the test ball and the sliding motion. However, it was analyzed that the multiwall tube-shape structure consisting of graphene was deformed from outside the structure through contact pressure, and graphene flakes started to fall off, resulting in breakage of the covalent bonds. The damaged MWCNTs were degraded to a two-dimensional hexagonal graphene structure such as xGnPs, and it
Materials 2021, 14, x FOR PEER REVIEW 8 of 20
is assumed that a kind of graphene-based tribofilm was ultimately formed. The evidence for the deformation and lubrication mechanism of MWCNTs was found from several tribological studies. Reinert et al. proposed a structural degradation model of the mechan- ically stressed MWCNTs by using the custom-made ring-on-block tribometer [31]. Sakka et al. investigated the friction and wear of bulk epoxy and carbon filler reinforced epoxy composites using the pin-on-disc tribometer and presented the schematic diagram of structural transformation of MWCNTs [32]. As discussed in the previous section, the surface morphologies that seem to be added layers are observed in Figure 6b–e, and they could play roles of tribofilm during the ball on plate test. Therefore, the proposed lubrication mechanism given in Figure 7 could be Materials 2021, 14, 685 confirmed. 7 of 19 Meanwhile, as indicated by the test results, the hBN case had worse wear values than the dry case. In this case, there existed ceramic materials, hBN, and carbonous debris cre- ated by the friction test. As the two materials mixed with each other, the lubricity was 2.4.analyzed Discussions to be lower than that of the dry case by making three-body contact instead of creatingIn these tribofilm. tribological Therefore, test results, it was theconfirmed three graphene that there nanoplatelets was no lubrication (xGnP C-750, effect xGnPas the M-5,solid and lubricants, xGnP H-5) resulting and the in multiwallsubstantially carbon more nanotubes wear than (MWCNTs) in the dry couldcase. meaningfully reduceAs the shown coefficient by the oftest friction results and on friction the wear and width wear, and the depth.cases of However, the three hexagonalxGnPs and boronthe MWNCTs nitride (hBN) could was improve not effective. the lubrication To elucidate behavior the primary compared lubrication to the mechanismdry case while for eachforming case graphene-based reasonably, theFE-SEM tribofilm images by reducing given inthe Figure direct6 contactwere more between deeply the investigated ball and the byCFRP referencing specimen. some Moreover, studies in on these tribological cases, the characteristics effect of the ofself-lub carbonousricating materials. property Thewas analyzedmaximized schematic due to the expressions graphene-based are provided nanoparticles. in Figure 7.
FigureFigure 7.7. PrimaryPrimary lubricationlubrication mechanismmechanism inin eacheach casecase duringduring thethe ballball onon plateplate test.test.
3. Micro-DrillingIt has been known Experiments that carbonous with Nano-Solid materials basically Lubrication have the self-lubricating property. SureshaTo etbenefit al. reported from the a self-lubricating enhanced tribological nature of effect carbon of fibercarbonous from the nano-scale abrasion particles, test [27]. Zhuthree et xGnP al. conducted particles thewere pin chosen joint test for ofthe graphite nano-solid and dry carbon lubrication fiber modified of micro-drilling polymer, and ex- theperiments. self-lubricating These particles carbonous had characteristics a shape of two- weredimensional found by thin observing sheet, and worn it was surfaces expected and wear debris [28]. In addition, these carbonous materials could create tribofilm. Zhao et al. found that short carbon fibers improve the anti-wear properties of polymer composites during the block-on-ring test by forming tribofilm [29]. Luo et al. found the formation of transfer film of polymer-carbon fiber composite by acting dry-sliding during the block-on-
ring test [30]. Therefore, in the dry case, wear debris occurred by the breakage of the carbon fiber which composes the MD-CFRP specimen could stay between the two counterparts. These carbonous materials could have formed carbon-based tribofilm due to the repeated stress- ing cycles. As a result, dry-sliding behavior occurred due to the self-lubricating capability of carbonous materials in MD-CFRP, despite the lack of nano-solid lubricants. In this context, the carbonous nano-solid lubricants could enhance such lubricating capability even more. In the cases of the three xGnPs, it was analyzed that the very thin two-dimensional graphene layers (general thickness of single-layer graphene: ~0.34 nm) slid on each other by enabling proper lubrication action. The lubrication characteristics of xGnP C-750, xGnP M-5, and xGnP H-5 were difficult to distinguish from each other, Materials 2021, 14, 685 8 of 19
as shown in the test results. This could have been because the particles were rubbed by the continuous movement of reciprocating sliding while loading the normal force on the test ball, so analogous graphene-based tribofilm could be formed in all xGnP cases by increasing the contact pressure between the test ball and the specimen. In the case of MWCNTs, the tube-shape structure might have been maintained at the beginning of the lubrication phase, despite the load on the test ball and the sliding motion. However, it was analyzed that the multiwall tube-shape structure consisting of graphene was deformed from outside the structure through contact pressure, and graphene flakes started to fall off, resulting in breakage of the covalent bonds. The damaged MWCNTs were degraded to a two-dimensional hexagonal graphene structure such as xGnPs, and it is assumed that a kind of graphene-based tribofilm was ultimately formed. The evidence for the deformation and lubrication mechanism of MWCNTs was found from several tribological studies. Reinert et al. proposed a structural degradation model of the mechanically stressed MWCNTs by using the custom-made ring-on-block tribometer [31]. Sakka et al. investigated the friction and wear of bulk epoxy and carbon filler reinforced epoxy composites using the pin-on-disc tribometer and presented the schematic diagram of structural transformation of MWCNTs [32]. As discussed in the previous section, the surface morphologies that seem to be added layers are observed in Figure6b–e, and they could play roles of tribofilm during the ball on plate test. Therefore, the proposed lubrication mechanism given in Figure7 could be confirmed. Meanwhile, as indicated by the test results, the hBN case had worse wear values than the dry case. In this case, there existed ceramic materials, hBN, and carbonous debris created by the friction test. As the two materials mixed with each other, the lubricity was analyzed to be lower than that of the dry case by making three-body contact instead of creating tribofilm. Therefore, it was confirmed that there was no lubrication effect as the solid lubricants, resulting in substantially more wear than in the dry case. As shown by the test results on friction and wear, the cases of the three xGnPs and the MWNCTs could improve the lubrication behavior compared to the dry case while forming graphene-based tribofilm by reducing the direct contact between the ball and the CFRP specimen. Moreover, in these cases, the effect of the self-lubricating property was maximized due to the graphene-based nanoparticles.
3. Micro-Drilling Experiments with Nano-Solid Lubrication To benefit from the enhanced tribological effect of carbonous nano-scale particles, three xGnP particles were chosen for the nano-solid dry lubrication of micro-drilling experiments. These particles had a shape of two-dimensional thin sheet, and it was expected that they could significantly improve lubrication in micro-drilling of a CFRP workpiece with their inherent sliding action.
3.1. Experimental Set-Up In this research, a multi-directional CFRP (MD-CFRP) composite was used as the work- piece for micro-drilling process. The size of the workpiece was 30 mm × 30 mm × 3 mm. The MD-CFRP workpiece was fabricated with 11-ply in the directions of 0◦ and 90◦. The stacking sequence of the MD-CFRP composite was [90◦/0◦/0◦/0◦/90◦/90◦/90◦/0◦/0◦/0◦/ ◦ 90 ]s, and the carbon fiber of prepreg used in this workpiece was the same one used for the tribological test (Table2 in Section 2.2). To apply the nano-solid dry lubrication technique, a nano-solid lubrication spray module was designed and is shown in Figure8a. The module has an air inlet and an air outlet. A spiral-shape tube was used in an attempt to decrease the aggregation of the nano-solid lubricants, considering the cohesion of particles. For lubrication action, the air- driven lubricants were sprayed to the cutting zone via an external nozzle. The customized lubrication module was integrated to construct the entire nano-solid lubrication system. An air compressor (PAB 42T3-503F, Power Air, Yangpyeong-gun, Korea), a drain filter (WD-02, Materials 2021, 14, x FOR PEER REVIEW 9 of 20
that they could significantly improve lubrication in micro-drilling of a CFRP workpiece with their inherent sliding action.
3.1. Experimental Set-Up In this research, a multi-directional CFRP (MD-CFRP) composite was used as the workpiece for micro-drilling process. The size of the workpiece was 30 mm × 30 mm × 3 mm. The MD-CFRP workpiece was fabricated with 11-ply in the directions of 0° and 90°. The stacking sequence of the MD-CFRP composite was [90°/0°/0°/0°/90°/90°/90°/0°/0°/0°/ 90°]s, and the carbon fiber of prepreg used in this workpiece was the same one used for the tribological test (Table 2 in Section 2.2). To apply the nano-solid dry lubrication technique, a nano-solid lubrication spray module was designed and is shown in Figure 8a. The module has an air inlet and an air outlet. A spiral-shape tube was used in an attempt to decrease the aggregation of the nano- Materials 2021, 14, 685 solid lubricants, considering the cohesion of particles. For lubrication action, the 9air- of 19 driven lubricants were sprayed to the cutting zone via an external nozzle. The customized lubrication module was integrated to construct the entire nano-solid lubrication system. An air compressor (PAB 42T3-503F, Power Air, Yangpyeong-gun, Korea), a drain filter (WD-02,Win &Tec Win Korea, & Tec Anyang-si, Korea, Anyang-si, Korea), a Korea), precision a regulatorprecision (IRregulator 1000-01G, (IR SMC1000-01G, Coporation, SMC Coporation,Tokyo, Japan), Tokyo, and Japan), an air and duct an were air duct set to were help set spray to help the spray nano-solid the nano-solid dry lubricants, dry lub- as ricants,shown as in shown Figure 8inb. Figure 8b.
(a) 3D image of lubrication module (b) Entire nano-solid dry lubrication system
FigureFigure 8. 8.Nano-solidNano-solid lubrication lubrication module module and and schematic schematic diagram diagram of entire of entire nano-solid nano-solid dry drylubrica- lubrica- tiontion system. system.
TheThe three-axis three-axis horizontal horizontal machine machine tool tool sy systemstem with with a anano-solid nano-solid dry dry lubrication lubrication modulemodule is is shown shown in in Figure Figure 9,9 ,and and several several components components with with three three DC-motor DC-motor driven driven linear linear slidesslides (MX80S, (MX80S, Parker-Hannifin Parker-Hannifin Corporation, Corporation, Mayfield Mayfield Heights,Heights, OH,OH, USA),USA), electric electric spindle spin- dle(E-800Z, (E-800Z, Nakanishi, Nakanishi, Kanuma, Kanuma, Japan) Japan) and and programmable programmable multi-axis multi-axis controller controller (CEM (CEM 104, 104,Delta Delta Tau Tau Data Data Systems, Systems, Chatsworth, Chatsworth, CA, USA)CA, USA) were were prepared. prepared. An external An external spray spray nozzle nozzlewith anwith internal an internal diameter diameter of 3 of mm 3 mm was was set set 17 17 mm mm away away from from the the workpiece workpiece with with a spraying elevation angle of 60◦ for nano-solid dry lubrication. A two-flute uncoated a spraying elevation angle of 60° for nano-solid dry lubrication. A two-flute uncoated Materials 2021, 14, x FOR PEER REVIEWtungsten carbide drill (DIXI 1138, Dixi Polytool, Le Locle, Switzerland) with a diameter10 of of 20 tungsten carbide drill (DIXI 1138, Dixi Polytool, Le Locle, Switzerland) with a diameter of 800 µm was mounted to the electric spindle. 800 μm was mounted to the electric spindle.
FigureFigure 9.9. ExperimentalExperimental set-upset-up forfor MD-CFRPMD-CFRP machiningmachining usingusing nano-solidnano-solid drydry lubrication.lubrication
3.2.3.2. ExperimentalExperimental DesignDesign andand ConditionsConditions InIn order to to improve improve the the micro-drilling micro-drilling performance performance of ofMD-CFRP, MD-CFRP, three three graphene graphene na- nanoplateletsnoplatelets (xGnP) (xGnP) were were used used as as nano-lubricant nano-lubricants:s: xGnP xGnP C-750, C-750, xGnP xGnP M-5, M-5, and xGnP H-5. TheThe propertiesproperties of of xGnPs xGnPs are are summarized summarized in Figurein Figure1 and 1 Tableand Table1 in Section 1 in Section 2.1. In total,2.1. In there total, there were five experimental cases, including dry and pure air cases. In each case, a new tool and workpiece was used, and 92 instances of through-hole drilling were conducted with a hole depth of 3 mm. The drilling parameters, being the spindle speed, feed per revolution that were used in this paper were carefully chosen by considering the recom- mended drilling conditions of the manufacturer of the drill [33]. As a result, the spindle speed and feed per revolution were fixed at 20,000 RPM and 14 μm/rev, respectively, after several preliminary drilling experiments. In addition, the air pressure was 0.4 bar for the pure air and xGnPs nano-solid lubrication.
4. Micro-Drilling Performance Evaluation–Methods and Analysis To evaluate machining performances of the micro-drilling of MD-CFRP based on nano-solid dry lubrication using three graphene nanoplatelets, the general machining per- formance factors of CFRP composite drilling were considered [34–37]. The experimental results were comprehensively discussed in terms of surface quality (delamination, uncut fiber and inner surface) and tool wear.
4.1. Delamination To analyze the quality of outside of the drilled holes, the delamination at an exit hole was considered. The delamination can be occurred by the push-down mechanism. When the drilling thrust force exceeds the interlaminar bond strength before the CFRP laminate is totally drilled by the tool, this push-down can occur and result in the delamination. Thus, the delamination factor (Fd) explains the degree of compressive thrust force that the drill bit can penetrate on the CFRP [38]. The factor can be calculated by taking the ratio of the maximum diameter, Dmax, with respect to the nominal diameter, Dnom, of the micro-drilled hole. In the experiments, the maximum diameter, Dmax, is the maximum extension of delamination outside the drilled hole and the nominal diameter, Dnom, is equal to the tool diameter of 800 μm. To analyze
Materials 2021, 14, 685 10 of 19
were five experimental cases, including dry and pure air cases. In each case, a new tool and workpiece was used, and 92 instances of through-hole drilling were conducted with a hole depth of 3 mm. The drilling parameters, being the spindle speed, feed per revolution that were used in this paper were carefully chosen by considering the recommended drilling conditions of the manufacturer of the drill [33]. As a result, the spindle speed and feed per revolution were fixed at 20,000 RPM and 14 µm/rev, respectively, after several preliminary drilling experiments. In addition, the air pressure was 0.4 bar for the pure air and xGnPs nano-solid lubrication.
4. Micro-Drilling Performance Evaluation–Methods and Analysis To evaluate machining performances of the micro-drilling of MD-CFRP based on nano-solid dry lubrication using three graphene nanoplatelets, the general machining performance factors of CFRP composite drilling were considered [34–37]. The experimental results were comprehensively discussed in terms of surface quality (delamination, uncut fiber and inner surface) and tool wear.
4.1. Delamination To analyze the quality of outside of the drilled holes, the delamination at an exit hole was considered. The delamination can be occurred by the push-down mechanism. When the drilling thrust force exceeds the interlaminar bond strength before the CFRP laminate is Materials 2021, 14, x FOR PEER REVIEW 11 of 20 totally drilled by the tool, this push-down can occur and result in the delamination. Thus, the delamination factor (Fd) explains the degree of compressive thrust force that the drill bit can penetrate on the CFRP [38]. the delamination,The factor can a 3D be calculatedlaser microscope by taking (VK-X the series, ratio ofKeyence, the maximum Osaka, Japan) diameter, and D pro-max, cessingwith respect software to the were nominal used. diameter,Figure 10a D showsnom, of an the image micro-drilled of a sample hole. drilled In the hole experiments, with Dnom andthe maximumDmax for calculating diameter, the D maxdelamination, is the maximum factor. extension of delamination outside the drilledFor hole the andanalysis the nominalof each experimental diameter, Dnom case,, is equaldelamination to the tool factors diameter were of compared 800 µm. Toat theanalyze same the drilled delamination, hole. The delamination a 3D laser microscope factors for (VK-X 12 holes–the series, Keyence, 5th, 15th, Osaka, 23rd, Japan)28th, 38th, and 46th,processing 51st, 61st, software 69th, were74th, used.84th, and Figure 92nd 10 aholes shows out an of image 92 were of acalculated sample drilled at the holeexit hole, with andDnom theirand average Dmax for values calculating are shown the delamination in Figure 10b. factor.
(a) Sample image of measuring (b) Average delamination factor delamination factor FigureFigure 10. 10. SampleSample images images for for calculating calculating delamina delaminationtion and average delamination factor.
AsFor shown the analysis in Figure of each10b, the experimental average delami case,nation delamination factor in factorsthe dry werecase was compared substan- at tiallythe same higher drilled than hole. those The of delaminationthe other cases. factors Meanwhile, for 12 holes–the pure air, 5th, xGnP 15th, C-750, 23rd, xGnP 28th, M-5, 38th, and46th, xGnP 51st, 61st,H-5 nano-solid 69th, 74th, lubrication 84th, and 92nd could holes reduce out ofthe 92 delamination were calculated of the at theexit exit hole. hole, In particular,and their average xGnP M-5 values could are shownbe the inbest Figure option 10b. for decreasing the delamination factor, whileAs xGnP shown H-5 in also Figure seems 10b, to the be averageeffective delamination for reducing factorthe delamination in the dry case factor. was In substan- addi- tion,tially pure higher air also than could those reduce of the otherthe delamination cases. Meanwhile, factor over pure that air, of xGnP the dry C-750, case. xGnP Therefore, M-5, itand was xGnP confirmed H-5 nano-solid that nano-solid lubrication lubrication could reduceusing graphene the delamination nanoplatelets of the could exit hole.signifi- In cantly enhance the machining performance of MD-CFRP composite in terms of the reduc- tion of the delamination factor.
4.2. Uncut Fiber The mechanism of delamination and uncut fiber are interdependent. To investigate the quality of inside of the drilled holes, the uncut fiber area was studied at exit hole. The uncut fiber was generated more frequently in certain fiber cutting angles. It was known that the uncut fiber was generated at cutting angles between 90–180°, as the protrusion direction of the fibers parallel to the cutting edge, tool could cut the CFRP resin more easily than the fibers which has a higher tensile strength. In particular, at a cutting angle of 135°, the maximum uncut fiber usually occurs. On the other hand, at cutting angles between 0–90°, the cutting is appropriately done by the shear stress as the force is applied to the side of fiber when the tool rotates clockwise [39]. Figure 11a shows the image of measuring the uncut fiber area, and the cutting angles between 90–180° were expressed in a yellow color, observing uncut fiber. The uncut fiber area was estimated in this study using the image processing software (VK Analyzer, Osaka, Japan). The uncut fiber area, Af, is the difference between the nominal drilled-hole area, Anom, and the extracted area through image processing, Aex; Anom is the area of a circle with a diameter of 800 μm. Aex for calculating the uncut fiber area was extracted as a red color through image processing. The processing parameter was set with the tolerance of the RGB data of 10. The values of the calculated uncut fiber area for 12 holes were averaged, which was the same hole used for the delamination factor, and these are graphically provided in Figure 11b.
Materials 2021, 14, 685 11 of 19
particular, xGnP M-5 could be the best option for decreasing the delamination factor, while xGnP H-5 also seems to be effective for reducing the delamination factor. In addition, pure air also could reduce the delamination factor over that of the dry case. Therefore, it was confirmed that nano-solid lubrication using graphene nanoplatelets could significantly enhance the machining performance of MD-CFRP composite in terms of the reduction of the delamination factor.
4.2. Uncut Fiber The mechanism of delamination and uncut fiber are interdependent. To investigate the quality of inside of the drilled holes, the uncut fiber area was studied at exit hole. The uncut fiber was generated more frequently in certain fiber cutting angles. It was known that the uncut fiber was generated at cutting angles between 90–180◦, as the protrusion direction of the fibers parallel to the cutting edge, tool could cut the CFRP resin more easily than the fibers which has a higher tensile strength. In particular, at a cutting angle of 135◦, the maximum uncut fiber usually occurs. On the other hand, at cutting angles between 0–90◦, the cutting is appropriately done by the shear stress as the force is applied to the side of fiber when the tool rotates clockwise [39]. Materials 2021, 14, x FOR PEER REVIEW 12 of 20 Figure 11a shows the image of measuring the uncut fiber area, and the cutting angles between 90–180◦ were expressed in a yellow color, observing uncut fiber. The uncut fiber area was estimated in this study using the image processing software (VK Analyzer, Osaka, AsJapan). expected, The the uncut uncut fiber fiber area, area A fvalue, is the in difference the dry case between was the the highest nominal when drilled-hole compared area, withAnom those, and of theother extracted cases. Similar area through to the resu imagelt of processing,analyzing the Aex delamination;Anom is the factor, area of it a was circle foundwith that a diameter nano-solid of 800 lubricationµm. Aex couldfor calculating reduce the the uncut uncut fiber fiber area area more was than extracted the sole as air a red supplycolor case. through Meanwhile, image processing. the smallest The value processing was confirmed parameter in the was case set withof xGnP the toleranceH-5. This of suggeststhe RGB that data xGnP of 10. H-5 is more effective for reducing uncut fiber area than xGnP C-750.
(a) Image of measuring uncut fiber area (b) Average uncut fiber area
FigureFigure 11. 11.SampleSample images images for formeasuring measuring uncut uncut fiber fiber and and average average uncut uncut fiber fiber area. area.
ToThe qualitatively values of thecompare calculated the quality uncut fiberof the area drilled for 12holes, holes optical were averaged,images of whichthe 28th was holethe were same taken hole in used the case for the of four delamination different types factor, of and lubrication, these are and graphically these are providedshown in in FigureFigure 12. 11 Inb. the As case expected, of pure the air uncut without fiber graphene area value nanoplatelets, in the dry case remarkable was the highest delamina- when tioncompared and uncut with fiber those defects of other were cases. observed Similar at the to thedrilled result hole. of analyzingOn the other the delaminationhand, with xGnPfactor, M-5 it and was xGnP found H-5, that the nano-solid overall surface lubrication quality could was reducesignificantly the uncut improved fiber areain each more case.than It is the assumed sole air that supply the case.larger Meanwhile, graphene nanoplatelets the smallest (xGnP value wasM-5 confirmedand xGnP H-5) in the could case of helpxGnP efficient H-5. Thiscutting suggests action that at the xGnP cutting H-5 isan moregles between effective 90–180° for reducing by reducing uncut fiber friction area thanat thexGnP tool-workpiece C-750. interface, and by keeping the tool edge sharp through lubrication effect. As a result,To qualitatively the uncut fiber compare could thebe decreased. quality of the drilled holes, optical images of the 28th hole were taken in the case of four different types of lubrication, and these are shown in Figure 12. In the case of pure air without graphene nanoplatelets, remarkable delamination and uncut fiber defects were observed at the drilled hole. On the other hand, with xGnP M-5 and xGnP H-5, the overall surface quality was significantly improved in each case. It is assumed that the larger graphene nanoplatelets (xGnP M-5 and xGnP H-5) could help
(a) Air (b) xGnP C-750 (c) xGnP M-5 (d) xGnP H-5 Figure 12. Images of the 28th drilled exit hole of MD-CFRP in each experimental case.
4.3. Inner Surface In order to investigate the MD-CFRP hole quality in further detail, the drilled holes were cut in half using a diamond wheel cutter, and sectioned parts of the workpieces were analyzed. To analyze the drilled hole surface, schematic views of the sectioned hole are shown in Figure 13a. As mentioned above, MD-CFRP with two carbon fiber stacking di- rections (0°, 90°) was used in this study. Based on the cutting mechanism of CFRP com- posites, extensive bending-induced defects would be expected to occur at the fiber cutting angle of 135°. On the other hand, with the fiber cutting angles of 45° and 90°, the carbon fiber cutting is done by the shear stress that occurs as the force is applied to the side of fiber. In the case of the fiber cutting angle of 0°, the cutting is caused by the force on the fiber being lifted by the tool [40]. Therefore, in cases with the fiber cutting angles of 45° and 90°, the carbon fiber would be adequately cut compared to that with the fiber cutting angle of 135°. Considering the general cutting mechanism of CFRP, inner surface defects of the drilled hole with a fiber stacking direction of 90° might be found on the left side, where the fiber cutting angle is 135°. By the same reasoning, in the carbon fiber stacking
Materials 2021, 14, x FOR PEER REVIEW 12 of 20
As expected, the uncut fiber area value in the dry case was the highest when compared with those of other cases. Similar to the result of analyzing the delamination factor, it was found that nano-solid lubrication could reduce the uncut fiber area more than the sole air supply case. Meanwhile, the smallest value was confirmed in the case of xGnP H-5. This suggests that xGnP H-5 is more effective for reducing uncut fiber area than xGnP C-750.
(a) Image of measuring uncut fiber area (b) Average uncut fiber area Figure 11. Sample images for measuring uncut fiber and average uncut fiber area.
To qualitatively compare the quality of the drilled holes, optical images of the 28th hole were taken in the case of four different types of lubrication, and these are shown in Materials 2021, 14, 685 Figure 12. In the case of pure air without graphene nanoplatelets, remarkable delamina-12 of 19 tion and uncut fiber defects were observed at the drilled hole. On the other hand, with xGnP M-5 and xGnP H-5, the overall surface quality was significantly improved in each case. It is assumed that the larger graphene nanoplatelets (xGnP M-5 and xGnP H-5) could ◦ helpefficient efficient cutting cutting action action at the at cuttingthe cutting angles angles between between 90–180 90–180°by reducingby reducing friction friction at the at thetool-workpiece tool-workpiece interface, interface, and and by keepingby keeping the the tool tool edge edge sharp sharp through through lubrication lubrication effect. effect. As Asa result, a result, the the uncut uncut fiber fiber could could be be decreased. decreased.
Materials 2021, 14, x FOR PEER REVIEW 13 of 20
(a) Air (b) xGnP C-750 (c) xGnP M-5 (d) xGnP H-5 direction of 0°, defects could occur on the right side containing a cutting angle of 135°. In
FigureFigureFigure 13a,12. 12. ImagesImages nine sectorsof of the the 28th 28th ([a]–[i])drilled drilled in exit exit which hole hole ofthe of MD-CFRP MD-CFRP inner surface in in each each quality experimental experimental was investigated case. case. are shown. 4.3.4.3. Inner Inner Surface Surface To quantitatively analyze the inner surface quality, the arithmetical mean height (Ra) In order to investigate the MD-CFRP hole quality in further detail, the drilled holes was measuredIn order toby investigate obtaining 3Dthe profile MD-CFRP data holeof the quality 90th drilled in further hole detail,for each the case drilled using holes the were cut in half using a diamond wheel cutter, and sectioned parts of the workpieces were 3Dwere laser cut microscope in half using (VK-X a diamond series, wheel Keyence, cutter, Osaka, and sectioned Japan) and parts processing of the workpieces software were(VK analyzed. To analyze the drilled hole surface, schematic views of the sectioned hole are Analyzer,analyzed. Osaka, To analyze Japan). the The drilled device hole can surface, measur schematice fine shapes views more of a theccurately sectioned because hole the are shown in Figure 13a. As mentioned above, MD-CFRP with two carbon fiber stacking lasershown spot in diameter Figure 13a. (0.4 As μm) mentioned was much above, smaller MD-CFRP than a tip with radius two (2 μcarbonm) of afiber typical stacking contact- di- directions (0◦, 90◦) was used in this study. Based on the cutting mechanism of CFRP typerections stylus (0°, [41]. 90°) The was measurement used in this parameterstudy. Based of theon softwarethe cutting was mechanism set with a cut-offof CFRP length com- posites,composites, extensive extensive bending-induced bending-induced defects defects would wouldbe expected be expected to occur to at occur the fiber at the cutting fiber ofcutting 2.5 mm angle and a of measurement 135◦. On the length other hand,of 0.6 withmm for the nine fiber sectors, cutting according angles of to 45 the◦ and ISO 90 4287◦, the angle of 135°. On the other hand, with the fiber cutting angles of 45° and 90°, the carbon standard.carbon fiber The cuttingmeasured is done values by thewere shear averaged, stress thatand occurstheir results as the are force shown is applied in Figure to the 13b. side fiber cutting is done by the shear stress that occurs as the force is applied to the side of Overall,of fiber. the In thedeviation case of was the fiberlarge cuttingbecause angle the fiber of 0◦ cutting, the cutting angle is is caused different by in the each force sector. on the fiber. In the case of the fiber cutting angle of 0°, the cutting is caused by the force on the Infiber particular, being lifted the fiber by the cutting tool [ 40angle]. Therefore, at the sectors in cases of [c], with [d], the and fiber [i] cutting was 135°, angles and of the 45 ◦ fiber being lifted by the tool [40]. Therefore, in cases with the fiber cutting angles of 45° drilledand 90 surfaces◦, the carbon were fibermuch would rougher be adequatelydue to the cutbending compared mechanism to that withas explained the fiber in cutting the and 90°, the carbon fiber would be adequately cut compared to that with the fiber cutting cuttingangle ofmechanism 135◦. Considering of CFRP thegiven general in the cutting Section mechanism 4.2. of CFRP, inner surface defects of angle of 135°. Considering the general cutting mechanism of CFRP, inner surface defects theThe drilled average hole with surface a fiber roughness stacking directionvalue obtained of 90◦ might in the be dry found case on was the leftsubstantially side, where of the drilled hole with a fiber stacking direction of 90° might be found on the left side, higherthe fiber than cutting those anglein the is other 135◦ .cases. By the On same the reasoning,other hand, in the the values carbon decreased fiber stacking more direction when where the fiber cutting angle is 135°. By the same reasoning, in the carbon fiber stacking usingof 0◦ ,nano-solid defects could lubrication occur on with the right graphene side containing nanoplatelets a cutting than angle they ofdid 135 in◦. the In Figure pure air13a, case.nine Specifically, sectors ([a]–[i]) the smallest in which value the inner was surfaceconfirmed quality in the was xGnP investigated M-5 case. are shown.
(a) Schematic views of drilled hole (b) Average inner surface roughness and nine sectors ([a]–[i])
FigureFigure 13. 13. SchematicSchematic views views of of drilled drilled hole hole and and nine nine sectors sectors and and average average inner inner surface surface roughness. roughness.
FigureTo quantitatively 14 shows the analyze 3D profiles the innerthat were surface used quality, to analyze the arithmeticalthe inner hole mean surface height rough- (Ra) nesswas values measured for four by experimental obtaining 3D cases. profile In data the case of the of 90thpure drilledair, there hole are for some each defects, case using such as the fiber3D laserpull-out microscope at sectors (VK-X[d] and series, [i]. The Keyence, defects at Osaka, these sectors Japan) could and processingbe explained software by the cut- (VK tingAnalyzer, mechanism Osaka, of CFRP Japan). composite, The device as previous can measurely explained fine shapes in Figure more 13a. accurately The same defects because werethe laserfound spot at sectors diameter [c], (0.4[d], µandm) was[i] in much the xGnP smaller C-750 than case. a tip Meanwhile, radius (2 µ them) inner of a typical hole surfacecontact-type of the stylusxGnP [M-541]. Thecase measurementwas much smoot parameterher than of thethose software of the wasother set experimental with a cut-off cases.length In ofthe 2.5 case mm of and xGnP a measurement H-5, defects lengthwere reduced of 0.6 mm compared for nine sectors,to the air according case, but to the the surfaceISO 4287 appeared standard. to be The more measured uneven valuesthan the were xGnP averaged, M-5 case. and their results are shown in
Materials 2021, 14, 685 13 of 19
Figure 13b. Overall, the deviation was large because the fiber cutting angle is different in each sector. In particular, the fiber cutting angle at the sectors of [c], [d], and [i] was 135◦, and the drilled surfaces were much rougher due to the bending mechanism as explained in the cutting mechanism of CFRP given in the Section 4.2. The average surface roughness value obtained in the dry case was substantially higher than those in the other cases. On the other hand, the values decreased more when using nano-solid lubrication with graphene nanoplatelets than they did in the pure air case. Specifically, the smallest value was confirmed in the xGnP M-5 case. Figure 14 shows the 3D profiles that were used to analyze the inner hole surface roughness values for four experimental cases. In the case of pure air, there are some defects, such as fiber pull-out at sectors [d] and [i]. The defects at these sectors could be explained by the cutting mechanism of CFRP composite, as previously explained in Figure 13a. The same defects were found at sectors [c], [d], and [i] in the xGnP C-750 case. Meanwhile, Materials 2021, 14, x FOR PEER REVIEWthe inner hole surface of the xGnP M-5 case was much smoother than those of the14 otherof 20
experimental cases. In the case of xGnP H-5, defects were reduced compared to the air case, but the surface appeared to be more uneven than the xGnP M-5 case.
(a) Air (b) xGnP C-750 (c) xGnP M-5 (d) xGnP H-5
FigureFigure 14. 14. 3D3D profiles profiles of of the the 90th 90th drilled drilled ho holele and and the the nine nine sectors sectors ([a]–[i]) ([a]–[i]).
ForFor more more detailed detailed analysis, analysis, FE-SEM FE-SEM imag imageses of of the the 90th 90th hole hole are are shown shown in in Figures 15 ◦ ◦ Figuresand 16 15. To and investigate 16. To investigate the carbon the fiber carbon cutting fiber angles cutting of angles 0 and of 90 0° ,and images 90°, images were taken were at takensectors at sectors [e] and [e] [h], and respectively, [h], respectively, which are which shown are inshown Figure in 13 Figurea. 13a. AsAs shown shown in in Figure Figure 15, 15 in, in the the air air case, case, small small debris debris like like carbonous carbonous materials materials was was not not foundfound on on the the surface, surface, but but some some particles particles were were found found in in the the xGnP xGnP cases. cases. In In Figure Figure 15b, 15b, a a lotlot of of tiny tiny particles particles (<1 (<1 μµm)m) could could be be visually visually identified, identified, an andd this this is is judged judged as as xGnP xGnP C-750 C-750 µ (actual(actual size: size: ~0.2 ~0.2 μm).m). In InFigure Figure 15c,d, 15c,d, thin thin 2D-sheet 2D-sheet particles particles were were found, found, which which seem seem to beto attributable be attributable to xGnP to xGnP M-5 M-5 and and xGnP xGnP H-5, H-5, respectively respectively (actual (actual size size of both of both particles: particles: 5 µ μ5m).m). Therefore, Therefore, it was it was confirmed confirmed that that graphe graphenene nanoplatelets nanoplatelets penetr penetratedated appropriately appropriately in in each case during the nano-solid dry lubrication. each case during the nano-solid dry lubrication. Figure 16 shows the inner surface at the carbon fiber cutting angle of 0◦. In Figure 16a, fiber breakage was observed, which consists of defects caused by force on the fiber being lifted by a tool at a fiber cutting angle of 0◦. In Figure 16b, as was the case with the fiber cutting angle of 90◦ shown in Figure 15b, tiny particles which were presumed to be xGnP C-750 were attached on the carbon fiber. In Figure 16c,d, it was not possible to distinguish graphene nanoplatelets from carbon dust caused by fiber breakage. FE-SEM image analysis was used to study the drilled surface morphologies according to the cutting angles of 0◦ and 90◦, and it was also confirmed that the nano-solid lubricants properly penetrated into the drilled hole.
(a) Air (b) xGnP C-750
(c) xGnP M-5 (d) xGnP H-5 Figure 15. FE-SEM images of carbon fiber cutting angle of 90° (× 1400)
Figure 16 shows the inner surface at the carbon fiber cutting angle of 0°. In Figure 16a, fiber breakage was observed, which consists of defects caused by force on the fiber being lifted by a tool at a fiber cutting angle of 0°. In Figure 16b, as was the case with the fiber
Materials 2021, 14, x FOR PEER REVIEW 14 of 20
(a) Air (b) xGnP C-750 (c) xGnP M-5 (d) xGnP H-5 Figure 14. 3D profiles of the 90th drilled hole and the nine sectors ([a]–[i])
For more detailed analysis, FE-SEM images of the 90th hole are shown in Figures 15 and 16. To investigate the carbon fiber cutting angles of 0° and 90°, images were taken at sectors [e] and [h], respectively, which are shown in Figure 13a. As shown in Figure 15, in the air case, small debris like carbonous materials was not found on the surface, but some particles were found in the xGnP cases. In Figure 15b, a lot of tiny particles (<1 μm) could be visually identified, and this is judged as xGnP C-750 (actual size: ~0.2 μm). In Figure 15c,d, thin 2D-sheet particles were found, which seem to Materials 2021, 14, 685 be attributable to xGnP M-5 and xGnP H-5, respectively (actual size of both particles:14 5 of 19 μm). Therefore, it was confirmed that graphene nanoplatelets penetrated appropriately in each case during the nano-solid dry lubrication.
(a) Air (b) xGnP C-750
Materials 2021, 14, x FOR PEER REVIEW 15 of 20
cutting angle of 90° shown in Figure 15b, tiny particles which were presumed to be xGnP C-750 were attached on the carbon fiber. In Figure 16c,d, it was not possible to distinguish graphene nanoplatelets from carbon dust caused by fiber breakage. FE-SEM image anal-
ysis was used to study the drilled surface morphologies according to the cutting angles of (c) xGnP M-5 (d) xGnP H-5 0° and 90°, and it was also confirmed that the nano-solid lubricants properly penetrated intoFigureFigure the 15. drilled 15.FE-SEMFE-SEM hole. images images of carbon of carbon fibe fiberr cutting cutting angle angle of 90° of 90(×◦ 1400)(× 1400).
Figure 16 shows the inner surface at the carbon fiber cutting angle of 0°. In Figure 16a, fiber breakage was observed, which consists of defects caused by force on the fiber being lifted by a tool at a fiber cutting angle of 0°. In Figure 16b, as was the case with the fiber
(a) Air (b) xGnP C-750
(c) xGnP M-5 (d) xGnP H-5 FigureFigure 16. FE-SEM 16. FE-SEM images images of carbon of carbon fibe fiberr cutting cutting angle angle of 0° of (× 0 ◦1400).(× 1400).
4.4.4.4. Tool Tool Wear Wear In orderIn order to evaluate to evaluate the themachining machining performance performance in micro-drilling in micro-drilling of MD-CFRP of MD-CFRP using using the thenano-solid nano-solid dry drylubrication, lubrication, tool wear tool wearvalues values were obtained were obtained and analyzed and analyzed using a usingcon- a confocal microscope (VK-X series, Keyence) after drilling 92 holes for each experimental focal microscope (VK-X series, Keyence) after drilling 92 holes for each experimental case. case. In Figure 17a, a tool wear index was marked with yellow lines for quantitative In Figure 17a, a tool wear index was marked with yellow lines for quantitative comparison. comparison. Tool wear values were obtained by calculating the difference between the Tool wear values were obtained by calculating the difference between the diameter of a diameter of a new tool and the worn drill diameter. The values indicate the occurrence of new tool and the worn drill diameter. The values indicate the occurrence of tool wear at tool wear at the cutting edge and the margin of the drill bit. Therefore, the tool wear for the cutting edge and the margin of the drill bit. Therefore, the tool wear for each experi- each experimental case could be explained as the magnitude of friction between the carbon mental case could be explained as the magnitude of friction between the carbon fiber and fiber and the micro-drill. the micro-drill. The calculated tool wear values in all experimental cases are given in Figure 17b. In the dry case, the value was even higher than those of the other cases. The tool wear value in the pure air case was also smaller than that in the dry case. The tool wear is the smallest in the case of xGnP M-5 among all cases tested. The nano-solid dry lubrication with xGnP H-5 could also be slightly effective for improving the tool wear of a micro-drill compared to the pure air case. As expected from the above results, xGnP M-5 nanoparticles, which are larger than xGnP C-750 particles, were more advantageous for reducing the tool wear. The enhanced lubrication using air spray with nanoparticles of xGnP nano-solid lubrica- tion could lead to reduced tool wear.
Materials 2021, 14, 685 15 of 19 Materials 2021, 14, x FOR PEER REVIEW 16 of 20
(a) Tool wear index obtained by (b) Calculated tool wear values confocal microscope images
FigureFigure 17. 17. ToolTool wear wear index index obtained obtained by by confocal confocal mi microscopecroscope images images and and tool tool wear wear values. values.
FE-SEMThe calculated images of tool the wear tool valuesflank position in all experimental of Figure 17a cases are areshown given in in Figure Figure 18. 17 Allb. In casesthe dryinvolved case, thespraying value wasair condition, even higher in thancommon. those In of the the cases other of cases. pure The air tooland wearxGnP value C- 750,in theadhered pure airsubstance case was was also found smaller at the than flank that face in the of drythe case.tool. TheIn addition, tool wear abrasive is the smallest wear marksin the were case ofdiscovered xGnP M-5 around among the all casesmargin tested. in the The xGnP nano-solid C-750 drycase. lubrication Meanwhile, with in xGnPthe casesH-5 of could xGnP also M-5 be and slightly xGnP effective H-5, the for state improving of the flank the tool face wear appeared of a micro-drill relatively compared clean. As to previouslythe pure airstated, case. larger As expected graphene from nanoplatel the aboveets results, could xGnPeffectively M-5 nanoparticles,reduce the friction which be- are tweenlarger the than CFRP xGnP composite C-750 particles, and the were tool. more advantageous for reducing the tool wear. The enhanced lubrication using air spray with nanoparticles of xGnP nano-solid lubrication could lead to reduced tool wear. FE-SEM images of the tool flank position of Figure 17a are shown in Figure 18. All cases involved spraying air condition, in common. In the cases of pure air and xGnP C-750, adhered substance was found at the flank face of the tool. In addition, abrasive wear marks were discovered around the margin in the xGnP C-750 case. Meanwhile, in the cases of xGnP M-5 and xGnP H-5, the state of the flank face appeared relatively clean. As previously stated, larger graphene nanoplatelets could effectively reduce the friction between the CFRP composite and the tool. (a) Air (b) xGnP C-750
(a) Air (b) xGnP C-750 (c) xGnP M-5 (d) xGnP H-5 Figure 18. FE-SEM images of tool flank after drilling 92 holes (× 1000)
4.5. Discussions In order to comprehensively investigate the machining performance of micro-drilling of MD-CFRP using nano-solid dry lubrication, delamination, uncut fiber, inner surface quality, and tool wear were analyzed. Overall, nano-solid dry lubrication using xGnP C- 750 improved machinability compared to the pure air case, but it was not as effective as
xGnP M-5 and xGnP H-5. In other words, the larger graphene nanoplatelets (xGnP M-5 and xGnP H-5)(c) xGnP were moreM-5 advantageous than (smallerd) xGnP particles H-5 (xGnP C-750). This could be explained in relation to how much direct contact can be effectively reduced between FigureFigure 18 18.. FEFE-SEM-SEM images images of of tool tool flank flank after after drilling drilling 92 92 holes holes (× (× 1000)1000).
1
Materials 2021, 14, 685 16 of 19
4.5. Discussions In order to comprehensively investigate the machining performance of micro-drilling of MD-CFRP using nano-solid dry lubrication, delamination, uncut fiber, inner surface quality, and tool wear were analyzed. Overall, nano-solid dry lubrication using xGnP C-750 improved machinability compared to the pure air case, but it was not as effective as xGnP M-5 and xGnP H-5. In other words, the larger graphene nanoplatelets (xGnP M-5 Materials 2021, 14, x FOR PEER REVIEW 17 of 20 and xGnP H-5) were more advantageous than smaller particles (xGnP C-750). This could be explained in relation to how much direct contact can be effectively reduced between the tool grain and MD-CFRP workpiece. As confirmed in Section2, it has been shown that xGnPsthe tool with grain a two-dimensionaland MD-CFRP workpiece. sheet structure As con lubricatefirmed in while Section reducing 2, it has the been direct shown contact that betweenxGnPs with two a counterparts two-dimensional by creating sheet structure graphene-based lubricate tribofilm while reducing duringthe thesliding direct contact action. betweenMeanwhile, two counterparts the applied by forcecreating was graphe 50 N inne-based the ball tribofilm on plate during test in Sectionthe sliding2, and action. all carbonousMeanwhile, particles the seemed applied to force be smashed was 50 N into in eventhe ball finer on ones. plate Therefore, test in Section the measured 2, and all coefficientscarbonous particles of friction seemed could beto similarbe smashed regardless into even of geometrical finer ones. characteristicsTherefore, the ofmeasured various xGnPs,coefficients as can of friction be seen could in Figure be similar4a. However, regardless in of thegeometrical micro-drilling characteristics experiments, of various the measuredxGnPs, as can thrust be seen forces in were Figure in 4a. the However, range between in the micro-drilling 5 N and 6 N,which experiments, is much the smaller meas- thanured 50thrust N. Therefore, forces were the in xGnP the range nano-scale between particles 5 N and might 6 N, maintain which is their much two-dimensional smaller than 50 shapesN. Therefore, and these the xGnP geometrical nano-scale characteristics particles might could maintain affect lubrication their two-dimensional behavior. Thus, shapes the lubricationand these geometrical mechanisms characteristics at the interface could between affect thelubrication tool and behavior. the workpiece Thus, the during lubrica- the micro-drillingtion mechanisms process at the were interface schematized between in the consideration tool and the of workpiece the actual during sizes of the the micro- three xGnPsdrilling and process the toolwere grain schematized (ultrafine in grade:consideration 0.3–0.5 ofµ m),the actual and these sizes schematicsof the three arexGnPs shown and inthe Figure tool grain 19. (ultrafine The points grade: of direct 0.3–0.5 contact μm), and between these theschematics tool grain are andshown the in MD-CFRP Figure 19. The are markedpoints of in direct red. contact between the tool grain and the MD-CFRP are marked in red.
FigureFigure 19.19. LubricationLubrication mechanism of micro-drilling of MD-CFRP using nano-solid dry lubrication lubrication accordingaccording toto thethe averageaverage particleparticle sizesize (APS)(APS) ofof graphenegraphene nanoplatelets.nanoplatelets
InIn thethe casescases ofof drydry andand purepure air,air, thethe tooltool graingrain andand MD-CFRPMD-CFRP mightmight contactcontact directlydirectly withoutwithout lubricationlubrication action,action, resultingresulting inin aa smallersmaller improvementimprovement ofof thethe machinability.machinability. TheThe xGnPxGnP M-5M-5 and and xGnP xGnP H-5 H-5 could could effectively effectively decrease decrease direct direct contact contact due due to theirto their larger larger particle par- sizeticle (5sizeµm) (5 thanμm) than that ofthat tool of graintool grain (0.3–0.5 (0.3–0.5µm). μ However,m). However, in the in experimentalthe experimental results, results, the differencethe difference of machining of machining performance performance in the in two the xGnPstwo xGnPs (xGnP (xGnP M-5 andM-5 xGnP and xGnP H-5) withH-5) differentwith different thicknesses thicknesses was notwas clear. not clear. The xGnP The xGnP C-750 C-750 with awith smaller a smaller size (~0.2 size µ(~0.2m) thanμm) toolthan grain tool grain would would not have not reducedhave reduced direct contactdirect contact as effectively as effectively as the largeras the xGnPslarger (xGnPxGnPs M-5(xGnP and M-5 xGnP and H-5). xGnP Therefore, H-5). Therefore, in the micro-drilling in the micro-drilling of MD-CFRP, of MD-CFRP, it was analyzed it was analyzed that the largerthat the graphene larger graphene nanoplatelets nanoplatelets could contribute could co tontribute more efficient to more lubricationefficient lubrication at the moment at the ofmoment machining. of machining.
5. Conclusions In this paper, the characteristics of micro-drilling of MD-CFRP were evaluated while considering the tribological behavior of graphene nanoplatelets as nano-solid lubricants. First, to confirm the tribological behavior of the nano-solid lubricants against CFRP com- posite, the ball on plate test was conducted to investigate the coefficient of friction, wear width, and wear depth. In addition to three xGnPs (xGnP C-750, xGnP M-5, and xGnP H- 5), MWCNTs and hBN were also selected for comparison in this tribological test. The test results showed that the three xGnPs and the MWCNTs, which consist of 2D graphene sheets could help improve the lubrication behavior by reducing the direct contact between the two counterparts (the test ball and the specimen) due to the creation of graphene-
Materials 2021, 14, 685 17 of 19
5. Conclusions In this paper, the characteristics of micro-drilling of MD-CFRP were evaluated while considering the tribological behavior of graphene nanoplatelets as nano-solid lubricants. First, to confirm the tribological behavior of the nano-solid lubricants against CFRP com- posite, the ball on plate test was conducted to investigate the coefficient of friction, wear width, and wear depth. In addition to three xGnPs (xGnP C-750, xGnP M-5, and xGnP H-5), MWCNTs and hBN were also selected for comparison in this tribological test. The test results showed that the three xGnPs and the MWCNTs, which consist of 2D graphene sheets could help improve the lubrication behavior by reducing the direct contact between the two counterparts (the test ball and the specimen) due to the creation of graphene-based tribofilm. On the other hand, hBN, which consists of ceramic materials, could not lessen friction as it prevented the creation of carbon-based tribofilm. Through the micro-drilling of MD-CFRP with the nano-solid dry lubrication, delami- nation, uncut fiber, inner surface quality, and tool wear were comprehensively investigated, and the lubrication mechanism in each experimental case was analyzed by reflecting the actual sizes of the nano-solid lubricants (xGnP nano-scale particles) and the tool grain according to scale. Among the three xGnPs, the larger xGnPs (xGnP M-5 and xGnP H-5) decreased surface defects and tool wear significantly, and the reduction rates were the following when compared to pure air case: Delamination—5.3%; uncut fiber—63.0%; inner surface roughness—31.7%; tool wear—18.5%. Meanwhile, the smaller one (xGnP C-750) was less effective, and the reduction rates were the following compared to pure air case: delamination—1.8%; uncut fiber—34.6%; inner surface roughness—15.5%; tool wear—3.7%. Altogether, to guarantee lubrication effects, it was concluded that the average particle size of the graphene nanoplatelets should be sufficiently larger to reduce the points of direct contact between the tool and the CFRP.
Author Contributions: Conceptualization, J.W.K. and S.W.L.; methodology, J.W.K. and S.W.L.; soft- ware, J.W.K. and J.J.; validation, J.W.K. and J.N.; formal analysis, J.W.K.; investigation, J.J.; resources, S.W.L.; data curation, J.J; writing—original draft preparation, J.W.K. and S.W.L.; writing—review and editing, J.W.K., J.N. and S.W.L.; visualization, J.W.K.; supervision, S.W.L.; project administration, J.N.; funding acquisition, J.N. All authors have read and agreed to the published version of the manuscript. Funding: This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) [No. 2019R1F1A1059555] and by the Technology Innovation Program (20013794, Center for Composite Materials and Concurrent Design) funded by the Ministry of Trade, Industry & Energy (MOTIE, Korea). Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: The data presented in this study are available on request from the corresponding author. Conflicts of Interest: The authors declare no conflict of interest.
Abbreviations
MD Multi-directional UD Uni-directional CFRP Carbon fiber reinforced plastic xGnPs Graphene nanoplatelets MWFs Metal working fluids Materials 2021, 14, 685 18 of 19
MQL Minimum quantity lubrication MWCNTs Multiwall carbon nanotubes COF Coefficient of friction hBN Hexagonal boron nitride FE-SEM Field emission scanning electron microscope APS Average particle size SSA Specific surface area COF Coefficient of friction FAW Fiber area weight ID Intensity of D band IG Intensity of G band I2D Intensity of 2D band IG Intensity of G band Fd Delamination factor Dnom Nominal diameter Dmax Maximum extension of delamination Af Uncut fiber area Anom Nominal drilled-hole area Aex Extracted area through image processing Ra Arithmetical mean height
References 1. Hegde, S.; Shenoy, B.S.; Chethan, K.N. Review on carbon fiber reinforced polymer (CFRP) and their mechanical performance. Mater. Today Proc. 2019, 19, 658–662. [CrossRef] 2. Shah, S.; Karuppanan, S.; Megat-Yusoff, P.; Sajid, Z. Impact resistance and damage tolerance of fiber reinforced composites: A review. Compos. Struct. 2019, 217, 100–121. [CrossRef] 3. Sugita, N.; Shu, L.; Kimura, K.; Arai, G.; Arai, K. Dedicated drill design for reduction in burr and delamination during the drilling of composite materials. CIRP Ann. 2019, 68, 89–92. [CrossRef] 4. Jia, Z.-Y.; Bai, Y.; Wang, F.-J.; Hao, J.-X. Effect of tool wear on drilling unidirectional CFRP laminates in different fiber cutting angles. Int. J. Adv. Manuf. Technol. 2020, 110, 89–99. [CrossRef] 5. Ismail, S.O.; Sarfraz, S.; Niamat, M.; Mia, M.; Gupta, M.K.; Pimenov, D.Y.; Shehab, E. Comprehensive Study on Tool Wear During Machining of Fiber-Reinforced Polymeric Composites. In Machining and Machinability of Fiber Reinforced Polymer Composites; Springer: Singapore, 2021; pp. 129–147. 6. Nam, J.S.; Lee, P.-H.; Lee, S.W. Experimental characterization of micro-drilling process using nanofluid minimum quantity lubrication. Int. J. Mach. Tools Manuf. 2011, 51, 649–652. [CrossRef] 7. Kim, J.S.; Kim, J.W.; Lee, S.W. Experimental characterization on micro-end milling of titanium alloy using nanofluid minimum quantity lubrication with chilly gas. Int. J. Adv. Manuf. Technol. 2017, 9, 2741–2749. [CrossRef] 8. Lee, P.-H.; Kim, J.W.; Lee, S.W. Experimental characterization on eco-friendly micro-grinding process of titanium alloy using air flow assisted electrospray lubrication with nanofluid. J. Clean. Prod. 2018, 201, 452–462. [CrossRef] 9. Almudaihesh, F.; Holford, K.; Pullin, R.; Eaton, M. The influence of water absorption on unidirectional and 2D woven CFRP composites and their mechanical performance. Compos. Part B Eng. 2020, 182, 107626. [CrossRef] 10. Wu, P.; Xu, L.; Luo, J.; Zhang, X.; Bian, W. Influences of long-term immersion of water and alkaline solution on the fatigue performances of unidirectional pultruded CFRP plate. Constr. Build. Mater. 2019, 205, 344–356. [CrossRef] 11. Xie, J.; Lu, Z.; Guo, Y.; Huang, Y. Durability of CFRP sheets and epoxy resin exposed to natural hygrothermal or cyclic wet-dry environment. Polym. Compos. 2019, 40, 553–567. [CrossRef] 12. Kerrigan, K.; Scaife, R.J. Wet vs dry CFRP drilling: Influence of cutting fluid on tool performance. Procedia CIRP 2018, 77, 315–319. [CrossRef] 13. Morden Machine Shop. Wet-Machining of Composites Crosses Boundaries. Available online: https://www.mmsonline.com/ articles/wet-machining-of-composites-crosses-boundaries (accessed on 16 January 2021). 14. Kim, J.W.; Nam, J.; Lee, S.W. Experimental study on micro-drilling of unidirectional carbon fiber reinforced plastic (UD-CFRP) composite using nano-solid lubrication. J. Manuf. Process. 2019, 43, 46–53. [CrossRef] 15. Le, H.; Howson, A.; Ramanauskas, M.; Williams, J.A. Tribological Characterisation of Air-Sprayed Epoxy-CNT Nanocomposite Coatings. Tribol. Lett. 2012, 45, 301–308. [CrossRef] 16. Moghadam, A.D.; Omrani, E.; Menezes, P.L.; Rohatgi, P.K. Mechanical and tribological properties of self-lubricating metal matrix nanocomposites reinforced by carbon nanotubes (CNTs) and grapheme—A review. Compos. Part B Eng. 2015, 77, 402–420. [CrossRef] Materials 2021, 14, 685 19 of 19
17. Berman, D.; Erdemir, A.; Sumant, A.V. Few layer graphene to reduce wear and friction on sliding steel surfaces. Carbon 2013, 54, 454–459. [CrossRef] 18. Yin, X.; Wu, F.; Chen, X.; Xu, J.; Wu, P.; Li, J.; Zhang, C.; Luo, J. Graphene-induced reconstruction of the sliding interface assisting the improved lubricity of various tribo-couples. Mater. Des. 2020, 191, 108661. [CrossRef] 19. XG Sciences. Available online: https://xgsciences.com/ko/product/xgnp/ (accessed on 15 January 2021). 20. Korea Nanomaterials. Available online: https://koreanano.co.kr/shop/item.php?it_id=KRU4500 (accessed on 15 January 2021). 21. Lower Friction. Available online: https://www.lowerfriction.com/ (accessed on 15 January 2021). 22. Quaranta, S.; Giorcelli, M.; Savi, P. Graphene and MWCNT Printed Films: Preparation and RF Electrical Properties Study. J. Nanomater. 2019, 2019, 1–9. [CrossRef] 23. Ferrari, A.C.; Meyer, J.C.; Scardaci, V.; Casiraghi, C.; Lazzeri, M.; Mauri, F.; Piscanec, S.; Jiang, D.; Novoselov, K.S.; Roth, S.; et al. Raman Spectrum of Graphene and Graphene Layers. Phys. Rev. Lett. 2006, 97, 187401. [CrossRef] 24. Calizo, I.; Balandin, A.A.; Bao, W.; Miao, F.; Lau, C.N. Temperature dependence of the Raman spectra of gra-phene and graphene multilayers. Nano Lett. 2007, 7, 2645–2649. [CrossRef] 25. Kim, C.-D.; Truong, N.T.N.; Pham, V.T.H.; Jo, Y.; Lee, H.-R.; Park, C. Conductive electrodes based on Ni–graphite core–shell nanoparticles for heterojunction solar cells. Mater. Chem. Phys. 2019, 223, 557–563. [CrossRef] 26. Toray Composite Materials America, Inc. Available online: https://www.toraycma.com/page.php?id=661 (accessed on 15 January 2021). 27. Suresha, B.; Chandramohan, G.; Samapthkumaran, P.; Seetharamu, S. Three-body abrasive wear behaviour of carbon and glass fiber reinforced epoxy composites. Mater. Sci. Eng. A 2007, 443, 285–291. [CrossRef] 28. Zhu, J.; Xie, F.; Dwyer-Joyce, R.S. PEEK Composites as Self-Lubricating Bush Materials for Articulating Revolute Pin Joints. Polymers 2020, 12, 665. [CrossRef][PubMed] 29. Zhao, F.; Gao, C.; Wang, H.; Wang, T.; Wetzel, B.; Jim, B.-C.; Zhang, G.; Wang, Q. Tribological Behaviors of Carbon Fiber Reinforced Epoxy Composites Under PAO Lubrication Conditions. Tribol. Lett. 2016, 62, 37. [CrossRef] 30. Luo, W.; Liu, Q.; Li, Y.; Zhou, S.; Zou, H.; Liang, M. Enhanced mechanical and tribological properties in polyphe-nylene sulfide/polytetrafluoroethylene composites reinforced by short carbon fiber. Compos. Part B Eng. 2016, 91, 579–588. [CrossRef] 31. Reinert, L.; Varenberg, M.; Mücklich, F.; Suárez, S. Dry friction and wear of self-lubricating car-bon-nanotube-containing surfaces. Wear 2018, 406, 33–42. [CrossRef] 32. Sakka, M.M.; Antar, Z.; Elleuch, K.; Feller, J.F. Tribological response of an epoxy matrix filled with graphite and/or carbon nanotubes. Friction 2017, 5, 171–182. [CrossRef] 33. DIXI Polytool. Tungsten Carbide and Diamond Precision Tools (Catalog). Available online: https://dixipolytool.ch//wp- content/uploads/special_tool/cat_18_complet_EN.pdf (accessed on 22 January 2021). 34. Tran, Q.-P.; Nguyen, V.-N.; Huang, S.-C. Drilling Process on CFRP: Multi-Criteria Decision-Making with Entropy Weight Using Grey-TOPSIS Method. Appl. Sci. 2020, 10, 7207. [CrossRef] 35. Barnes, S.; Bhudwannachai, P.; Dahnel, A.N. Drilling Performance of Carbon Fiber Reinforced Epoxy Composite When Machined Dry, with Conventional Cutting Fluid and With a Cryogenically Cooled Tool. In ASME International Mechanical Engineering Congress and Exposition; American Society of Mechanical Engineers: New York, NY, USA, 2013; Volume 56192, p. V02BT02A063. 36. Gaugel, S.; Sripathy, P.; Haeger, A.; Meinhard, D.; Bernthaler, T.; Lissek, F.; Kaufeld, M.; Knoblauch, V.; Schneider, G. A comparative study on tool wear and laminate damage in drilling of carbon-fiber reinforced polymers (CFRP). Compos. Struct. 2016, 155, 173–183. [CrossRef] 37. Voss, R.; Henerichs, M.; Rupp, S.; Kuster, F.; Wegener, K. Evaluation of bore exit quality for fibre reinforced plastics including delamination and uncut fibres. CIRP J. Manuf. Sci. Technol. 2016, 12, 56–66. [CrossRef] 38. Durão, L.M.P.; Tavares, J.M.R.S.; De Albuquerque, V.H.C.; Marques, J.F.S.; Andrade, O.N. Drilling Damage in Composite Material. Materials 2014, 7, 3802–3819. [CrossRef] 39. Kwon, B.-C.; Mai, N.D.D.; Cheon, E.S.; Ko, S.-L. Development of a step drill for minimization of delamination and uncut in drilling carbon fiber reinforced plastics (CFRP). Int. J. Adv. Manuf. Technol. 2019, 106, 1291–1301. [CrossRef] 40. Ramulu, M. Machining and surface integrity of fibre-reinforced plastic composites. Sadhana 1997, 22, 449–472. [CrossRef] 41. Keyence. Available online: https://www.keyence.com/ (accessed on 23 January 2021).