Synergy between boron nitride or graphene nanoplatelets and tri(butyl)ethylphosphonium diethylphosphate ionic liquid as lubricant additives of triisotridecyltrimellitate oil

1 Synergy between boron nitride or graphene 1 nanoplatelets and tri(butyl)ethylphosphonium 2 diethylphosphate ionic liquid as lubricant 3 additives of triisotridecyltrimellitate oil 4 José M. Liñeira del Río, Enriqueta R. López, Josefa Fernández* 5 Laboratory of Thermophysical Properties, Nafomat Group, Department of Applied Physics, 6 Faculty of Physics, University of Santiago de Compostela, 15782, Santiago de Compostela, 7 Spain 8 *Corresponding author. 9 E-mail address: [email protected] (J. Fernandez) 10 11 2 ABSTRACT: In this work, the synergy between an ionic liquid (IL) and nanoparticles as 12 additives of lubricants was studied. For this purpose, four dispersions based on graphene 13 nanoplatelets, GnPs, or nanoparticles of hexagonal boron nitride, h-BN, with or without the IL 14 tri(butyl) ethylphosphonium diethylphosphate in an ester type base oil, triisotridecyltrimellitate 15 (TTM), were prepared and tribologically analyzed as potential nanolubricants. The mass 16 concentration of the nanoadditives is 0.1 wt%, whereas for the IL it is 2 wt%. The prepared 17 blends were stable for three weeks. New density and viscosity values show that both properties 18 slightly increase with the addition of IL and/or nanoparticles. Tribological tests were performed 19 under a normal load of 20 N for TTM, the four dispersions and the TTM + 2 wt% IL mixture. 20 With respect to base oil, a maximum friction reduction of 33% was achieved for TTM/IL/GnP 21 nanodispersion. The best antiwear performance also corresponds to this same nanodispersion 22 with a wear track width reduction of 44% and a strong decrease of the average cross sectional 23 area of 65%, both respect to those obtained with the neat oil. In the case of wear scar depth, the 24 maximum reduction is 32% for TTM/IL/h-BN nanodispersion. In addition, the values for 25 roughness of worn surfaces tested with both TTM/IL/GnP and TTM/IL/h-BN nanodispersions 26 are lower than those corresponding to the neat oil, to the TTM/IL mixture and to those of the 27 corresponding binary dispersions. Hence, positive synergies between the IL and GnP or h-BN 28 as additives of TTM were found. Confocal Raman microscopy demonstrates tribofilm 29 formation and mending effect on worn surfaces. 30 31 KEYWORDS: ester; ionic liquid; lubricant; nanoadditives; friction: wear 32 33 3 1. Introduction 34 Recent calculations of the impact of friction and wear indicate that 23% (119 EJ) of the 35 total energy consumption in the world takes place in tribological contacts (lubricated or not 36 lubricated solid surfaces). 20% (103 EJ) of total consumption is used to overcome friction and 37 3% (16 EJ) to remanufacture worn parts and spare equipment due to wear and related faults [1]. 38 The use of nanotechnology in the development of more efficient lubricants will reduce not only 39 these expenses but also CO2 emissions. In fact, the addition of a very low quantity of 40 nanoparticles to lubricants can improve their tribological performance. Nanoparticle additives 41 have superior tribological properties to traditional solid lubricant additives [2]. Nanolubricants 42 are stable colloidal suspensions of nanometric materials with a very low concentration (usually 43 lower than 1 wt%) in conventional lubricants. Hexagonal boron nitride (h-BN) based 44 nanolubricants lead to better antifriction/antiwear capabilities compared to those of several base 45 oils [3-7]. h-BN, considered an environmentally friendly material [8], is the softest and most 46 lubricious polymorph of BN [9], having a lamellar crystalline structure in which van der Waals 47 forces exist between sheets [10]. Moreover, graphene nanoplatelets (GnP) have been studied as 48 nanoadditives [11-13] improving frictional and antiwear characteristics as well as extreme 49 pressure properties, compared to those of some base fluids. Chang and Baek [14] have recently 50 summarized the ecofriendly green synthesis procedures of GnP. 51 On the other hand, good tribological performance has been reported for ionic liquids 52 (IL) studied as neat lubricants [15-20]. Moreover, the use of nanoadditives for ionic liquids as 53 base oils can improve their tribological behavior [21,22]. However, ILs are still expensive and 54 the current direction is to focus the research on the use of ionic liquids as additives [23]. In this 55 vein, ILs based on phosphonium cations have shown a good performance as additives of 56 biodegradable oils, among other lubricants, for steel/steel contacts [23-26]. 57 4 Moreover, one of the main problems when using nanoparticles as additives is the poor 58 stability of the resulting nanodispersions. The use of dispersants or chemical functionalization 59 of the nanoparticles are solutions under consideration. Besides, by combining desired properties 60 of ionic liquids and of nanoparticles, better stability and higher efficiency can be achieved 61 [22,27,28]. Such hybrid formulations sometimes exhibit interesting positive synergies [29], but 62 the investigations on the combined effects of ILs and uncoated nanoparticles as oil additives 63 are still very scarce [30-33]. 64 Senatore et al. [30] studied dispersions of a polyalkylene glycol base oil with both 1-65 ethyl-3-methylimidazolium acetate and graphene oxide (GO) as additives at two temperatures 66 (298.15 K and 353.15 K) finding friction reductions up to 17% at 298.15 K whereas at 353.15 67 K no reduction was found. For the lowest GO concentration, at 298.15 K the wear reduction 68 was 22% while for 353.15 K no reduction was obtained. On the other hand, for the highest GO 69 concentration the wear reductions were excellent at both temperatures. Sanes et al. [31] found 70 that the presence of 1-octyl-3-methylimidazolium tetrafluoroborate enhances the load-carrying 71 and surface separating ability of graphene, leading to an unmeasurable wear, when both 72 additives are dispersed in an isoparaffinic base oil, whereas for a SAE 10W30 fully formulated 73 oil, no positive synergies were found for the same additives. However, tetrafluoroborate ILs are 74 not recommended in tribology applications due to their reactivity with water leading to the 75 production of corrosive hydrogen fluoride acid [34-36]. Amiril et al. [32] concluded that the 76 addition of the IL trihexyltetradecylphosphonium bis(2,4,4-trimethylpentyl)phosphinate (1 77 wt%) and nanoparticles of hexagonal boron nitride (0.05 wt%) in a chemically modified palm 78 olein trimethylolpropane ester, slightly decreases both wear (approximately 3.4%) and friction 79 (5%) with respect to the base oil. Finally, Li et al. [33] investigated the synergistic effects of 2-80 mercaptobenzothiazolate based ILs and Mo nanoparticles in a polyethylene glycol base oil 81 observing excellent friction-reduction and anti-wear performance at 100 °C, but not at 20 °C. 82 5 Hence, as no clear conclusion can be found, more studies on IL and nanoadditives synergies 83 are needed. 84 In order to gain a deeper knowledge on the combined effects of ILs and uncoated 85 nanoparticles as oil additives, in this work we have analyzed the synergies of 86 tri(butyl)ethylphosphonium diethylphosphate, [P4,4,4,2][C2C2PO4], with hexagonal boron nitride 87 (h-BN) nanoparticles and with graphene nanoplatelets (GnPs) using triisotridecyltrimellitate 88 (TTM) as base oil. There is no previous research on the tribological synergies of both graphene 89 nanoplatelets and ILs as additives of base oils. Recently, Oulego et al. [37] have determined the 90 bacterial toxicity of seven phosphonium ILs previously analyzed as lubricant additives. These 91 authors conclude that the IL tri(butyl)ethylphosphonium diethylphosphate, [P4,4,4,2][C2C2PO4], 92 was the least toxic of all the ILs tested. The main advantages of trimellitate esters are low 93 volatility, good thermal stability, good stability to oxidation, high film strength, good low 94 temperature properties, high flashpoints and good hydrolytic stability. Trimellitate esters are 95 used as specialty lubricants including compressor fluids, two stroke oils, greases or chain oils 96 [38]. 97 98 2. Experimental section 99 2.1. Materials 100 The triisotridecyltrimellitate sample (TTM, CAS Number: 72361-35-4, Fig. 1) was 101 provided by Verkol. This oil has been characterized by infrared spectroscopy (IR) with a FTIR 102 Varian 670-IR spectrometer. The spectrum (Fig. 2) shows the following peaks: a strong peak 103 at 1726 cm-1, which corresponds to the stretching vibration of ester carbonyl (C=O), two weak 104 peaks around 1574 and 1607 cm-1 that are associated with the C-H stretching in-of plane ring, 105 some peaks appear around 1305–1240 cm−1 related to C–O(H) stretching and C–O–(H) bending 106 vibrations, a peak at 1102 cm-1, which can be assigned to the (C–O–C) single bond stretching 107 6 vibration and some peaks at 2956 cm-1, 2927 cm-1 and 2871 cm-1 which correspond to carbon–108 hydrogen groups: (CH3) asymmetric stretching, (CH2) asymmetric stretching and (CH3) 109 symmetric stretching respectively [39,40]. 110 O O O O O O 111 112 Fig. 1. Chemical structure of triisotridecyltrimellitate (TTM). 113 114 Fig. 2. The FTIR spectrum of triisotridecyltrimellitate (TTM) base oil. 115 Moreover, this base oil was also analyzed by high performance liquid chromatography, 116 HPLC, coupled with a quadrupole orthogonal acceleration time-of-flight mass spectrometer 117 (microTOF-Q™) which is equipped with an electrospray ionization source (ESI). The base oil 118 was dissolved in isopropanol (5:250) and analyzed in isocratic mode. As can be seen in Fig. 3, 119 7 a wide peak appears, which may be due to small impurities similar to triisotridecyltrimellitate. 120 The mass spectrum (Fig. 4) of the oil, shows that the mass of the molecular ion corresponds to 121 TTM (molecular weight 757.63 g·mol-1 and molecular formula: C48H84O6). Moreover, this 122 spectrum shows another weak peak with a molecular weight close to that of TTM; this fact may 123 be owing to the loss of some hydrogen atoms in the molecule. To the best of our knowledge no 124 previous mass spectra of this type of esters have been reported. The TTM kinematic viscosity 125 at 40ºC and its viscosity index are 317 cSt and 74, respectively [24]. 126 127 Fig. 3. HPLC chromatogram of the triisotridecyltrimellitate sample. 128 129 Fig. 4. Mass spectrum of TTM (retention time 4.0 min). 130 Hexagonal boron nitride powders (h-BN, CAS Number: 10043-11-5) have a purity of 131 99.5 %, an average particle size of 70 nm and a specific average area of 19.4 m2/g as indicated 132 Intensity Time/min Intensity m/z 8 by the manufacturer (Iolitec, GmbH, Germany, lot MNC018001). An aliquot of the powder 133 sample has previously been characterized [41]. Disc-like shaped morphology was obtained for 134 h-BN nanoparticles with Transmission Electron Microscopy (TEM) [41]. Graphene 135 nanoplatelets powders (GnP, CAS number 1034343-98-0) of a purity of 99.5% with an average 136 particle diameter of 15 µm and a thickness of 11-15 nm, were also provided by Iolitec. An 137 aliquot of this GnP sample was used in a previous work [11] where its properties are described. 138 Tri(butyl)ethylphosphonium diethylphosphate ([P4,4,4,2][C2C2PO4], Cyphos 169, CAS 139 Number: 20445-94-7, Fig. 5) was kindly provided by Cytec Industries Inc. (US) with a purity 140 of 96.3%. Its kinematic viscosity at 40ºC and the viscosity index are 225 cSt and 82 [24], 141 respectively. FTIR and Raman spectra of this IL are shown in Figs. S1 and S2 respectively. A 142 band with three peaks at 2873, 2927 and 2958 cm-1 due to the methylene bonds of the aliphatic 143 chains can be observed in Fig. S1. FTIR spectrum also shows a single peak at 1148 cm-1 that 144 can be assigned to the P=O bond. The full spectrum is very similar to that previously reported 145 for the same IL by Hernández Battez et al. [42]. A WITec alpha300R+ confocal Raman 146 microscopy was used to obtain the Raman spectrum of the IL. We are not aware of any previous 147 Raman spectrum of this IL reported in the literature. 148 149 Fig. 5. Chemical structure of tri(butyl)ethylphosphonium diethylphosphate, 150 [P4,4,4,2][C2C2PO4]. 151 152 2.2. Preparation of the nanolubricants 153 Four dispersions were prepared apart from a mixture of TTM and [P4,4,4,2][C2C2PO4] 154 with a concentration of 2 wt% of the IL. Two-step method was used to make the 155 nanodispersions TTM + 0.1 wt% h-BN and TTM + 0.1 wt% GnP. To prepare TTM + 2 wt% 156 9 IL + 0.1 wt% h-BN and TTM + 2 wt% IL + 0.1 wt% GnPs nanodispersions, a procedure similar 157 to that used by Sanes et al. [31] was employed. Firstly, h-BN or GnP nanopowders were added 158 to the IL. Secondly, this ensemble is mechanically mixed in an agate mortar for 5 min and 159 subsequently mixed with the base oil (TTM). The four nanodispersions were sonicated for 4 160 hours by ultrasound in a Fisherbrand bath, operating in continuous shaking mode with an 161 effective power of 180 W and a sonication frequency of 37 kHz. The weight percentage of all 162 the blends was determined by using a Sartorius balance (model MC 210P) with a readability of 163 0.01 mg. The mass concentration of the nanoparticles was chosen due to good tribological 164 performance obtained in our previous research [11,43]. Stability of the nanodispersions were 165 analyzed by visual observation and the measurement of the refractive index along time by using 166 a Mettler Toledo Refractometer RA-510M. Its measuring cell is an inverted cone-shaped cavity, 167 with stainless steel walls. The base of this cone is a polished surface of a sapphire prism, on 168 which the nanolubricant is placed. In addition, in order to analyze the interactions among the 169 components of the nanodispersion the FTIR Varian 670-IR spectrometer was used. 170 2.3. Thermophysical Measurements 171 Density and dynamic viscosity of the lubricants were measured from 278.15 to 373.15 172 K and at atmospheric pressure with a rotational Stabinger viscometer SVM 3000 from Anton 173 Paar (Graz, Austria) which incorporates a vibrating tube densimeter [44]. This device has 174 previously been described in detail [45,46]. The expanded uncertainties (k = 2) are 1% for 175 dynamic viscosity, 0.0005 g⋅cm-3 for density and 0.02 K for the temperature from 288.15 to 176 378.15 K and 0.05 K outside this range. 177 2.4. Tribological Tests 178 Rotational friction tests were performed with a CSM Standard tribometer working in a 179 ball-on-disc configuration [43] for the base oil, the TTM + IL mixture and the four 180 nanodispersions at room temperature (~23ºC) under the following conditions: load of 20 N 181 16 Table 3. Mean values of the friction coefficient, μ, and of the width, WTW, depth, WTD, and 267 cross-section area of the wear track and their respective standard deviations for all lubricants. 268 Dispersion µ σ WTW/µm σ/µm WTD/µm σ/µm Area/µm2 σ/µm2 TTM 0.1253 0.0043 344 22 1.29 0.42 274 17 TTM+ 0.1 wt% h-BN 0.1076 0.0035 246 20 0.95 0.25 120 18 TTM+0.1 wt% GnP 0.1036 0.0060 242 21 1.23 0.19 142 12 TTM+ 2 wt% IL 0.1077 0.0096 234 18 1.21 0.37 140 14 TTM+2 wt% IL+ 0.1 wt% h-BN 0.1019 0.0043 223 19 0.88 0.51 120 12 TTM+2 wt% IL+ 0.1 wt% GnP 0.0836 0.0048 194 16 1.19 0.34 110 16 269 270 Fig. 9. Mean friction coefficient, µ , obtained for all the studied lubricants. 271 As can be seen in Table 3, the wear obtained with all the prepared dispersions is lower 272 than for the base oil without additives. The mean wear track width reduction ranges from 28% 273 (for the TTM/h-BN nanodispersion) to 44% (for the TTM/IL/GnP nanodispersion), the 274 improvements in the mean wear track depth range from 5% (for the TTM/GnP nanodispersion) 275 to 32% (for TTM/IL/h-BN nanodispersion) whereas the decrease of the average cross sectional 276 area (Fig. 10) is excellent for all the cases ranging from 56% (for the TTM/GnP mixture) to 277 66% (for TTM/IL/GnP nanodispersion). Consequently, the nanodispersions containing ILs 278 provide better tribological properties than the TTM/IL mixture and the corresponding 279 17 nanodispersion without IL. On the other hand, the friction and wear reductions of the 280 TTM/IL/h-BN nanodispersion (19% and 32% respectively) are significantly better than those 281 obtained by Amiril et al. [32] for an ester/IL/h-BN lubricant (5% and 3.4%) and those obtained 282 by Senatore et al. [30] at 298. 15 K for a glycol/IL/GO nanodispersion (17% and 22%). 283 284 Fig. 10. WTW and transversal area obtained for all the studied lubricants. 285 In Fig. 11 a significant reduction can be observed in the 3D profiles of the wear tracks 286 obtained lubricating the contact with the TTM/IL/GnP nanodispersion in comparison with that 287 corresponding to neat oil. Profiles of the transversal areas and 3D profiles of worn surfaces for 288 all the studied lubricants are shown in Fig. 12, where the reductions of the scars are clearly 289 observed, especially for the TTM + 2 wt% IL + 0.1 wt% GnP nanodispersion. 290 18 291 Fig. 11. 3D profiles (Confocal 10x) of the wear tracks of the discs lubricated with (a) TTM 292 base oil (b) nanolubricant formed by TTM + 2 wt% IL + 0.1 wt% GnP 293 294 Fig. 12. a) 3D Surface topography of wear tracks and b) Cross section profiles of wear tracks 295 for the all the studied lubricants at room temperature. 296 Different mechanisms have been identified to explain the role of nanoparticles as 297 lubricant additives. These mechanisms are classified in two different categories: direct effect 298 19 of the nanoparticle on the surface (ball bearing and tribofilm formation) and surface 299 enhancement effects (mending and polishing effects) [49,50]. As regards the role of ionic liquid, 300 the main mechanism is the formation of tribofilm. In order to better understand the mechanisms 301 underlying in the present samples, we have performed measurements of roughness, SEM and 302 Raman microscopy. 303 The roughness of the worn surface, Ra, was determined according to the standard ISO 304 4287, applying a Gaussian filter with a long wavelength cut-off of 0.25 mm. As presented in 305 Table 4, the roughness values of the worn surface, Ra, corresponding to TTM+IL, TTM+h-BN 306 and TTM+GnP are 8.8, 8.3 and 7.9 nm respectively. Taking into account that the roughness 307 value of the worn surface of TTM base oil was 19.7 nm, we can conclude that the presence of 308 the chosen IL produces the formation of protective tribofilms whereas the nanoadditives can 309 lead to mending, polishing or tribofilm formation effects. The smoothest surface corresponds 310 to the disc lubricated with the TTM/IL/GnP nanodispersion (Ra = 7.0 nm) followed by the one 311 lubricated with the mixture TTM/IL/h-BN (Ra =7.5 nm). Thus, positive synergies between the 312 IL and the both nanoadditives were found. 313 Table 4. Roughness parameter, Ra, of worn surfaces for the different analyzed nanolubricants. 314 Lubricant Ra / nm σ / nm Gaussian Filter / mm TTM base oil 19.7 1.2 0.25 TTM +h-BN 8.29 0.64 0.25 TTM + GnP 7.91 0.55 0.25 TTM +IL 8.82 0.52 0.25 TTM +IL+ h-BN 7.47 0.51 0.25 TTM +IL+ GnP 7.02 0.46 0.25 Moreover, SEM micrographs of wear tracks after lubrication with all the studied 315 lubricants based on TTM were performed. Fig. 13 (especially Fig. 13b) shows abrasive wear 316 scratches for the scar corresponding to the neat oil whereas for the scars corresponding to the 317 five additivated TTM oils, plastic deformation and smoother surfaces were found in agreement 318 with roughness values. Additionally, for the nanolubricant TTM/IL/GnP an important decrease 319 20 in wear scar width is observed in comparison with the base oil TTM as seen in Fig. 13a. These 320 results confirm the wear measurements analyzed with the 3D profilometer. 321 322 323 Fig. 13. SEM micrographs a) 700x and b) 5000x for the worn discs after tribological tests for 324 the studied nanolubricants. 325 100 µm 100 µm100 µm 100 µm 100 µm 100 µm 10 µm 10 µm 10 µm 10 µm 10 µm 10 µm b) a) 21 Elemental mapping and Raman spectra of the three additives (Figs. S2-S4) and of the 326 worn surfaces lubricated with TTM and with the five additivated TTM oils (Figs. S5-S8 and 327 14) were recorded with a confocal Raman microscope at a wavelength of 532 nm in order to 328 know the role that nanoparticles and the ionic liquid play in the reduction of surface wear of 329 discs. An important tribofilm is evidenced due to a significant presence of IL (blue color) in the 330 mapping of the worn surface lubricated with the TTM/IL mixture. (Fig. S6). On the other hand, 331 the Raman spectrum of the h-BN powders (Fig S4) exhibits a characteristic band at 1367 cm-1 332 [51] that is due to the E2g phonon mode, similar to the G band in graphene or graphene 333 derivatives [11]. The spots of boron nitride nanoparticles (Fig. S7) on the worn surface 334 lubricated with TTM/h-BN dispersion, as well as roughness reduction could indicate the 335 occurrence of mending effect. 336 The spectrum of the GnP nanopowders (Fig. S3) shows two characteristic bands around 337 1350 cm-1 (D-band) and at 1580 cm-1 (G-band) [11]. The first one is a result of the breathing 338 modes of sp2 atoms in rings whereas the G-band is due to the bond stretching of all pairs of sp2 339 atoms in rings and chains [11,52]. On the worn surface lubricated with TTM/GnP dispersion 340 (Fig. S8) the presence of areas where the spectrum coincides with that of GnP can be observed 341 (Fig S3). Taking this last result into account, as well as the roughness and SEM images it can 342 be concluded that there is a presence of mending and tribofilm effects. 343 Fig. 14a, corresponding to the nanolubricant TTM/IL/h-BN, shows the presence of the 344 IL (blue) and h-BN (green) on the worn surface. The three spectra in Fig. 14a agree with the 345 Raman spectrum of the neat IL (Fig. S2), with those of the worn surface lubricated with TTM 346 (Fig. S5) and of h-BN [41] nanopowders. The Raman analyses of TTM/IL/GnP on the scar 347 surface are shown in Fig. 14b where red and blue show the presence of GnP or TTM and IL, 348 respectively. In this figure, the carbon spectrum agrees with the spectrum of the worn surface 349 lubricated with TTM (Fig. S5) and that of the GnP nanopowders (Fig S3), presenting the peaks 350 22 corresponding to the D and G-bands, as well as an additional band (2D band) which indicates 351 the presence of GnP. Moreover, the IL spectrum of Fig. 14b coincides with that of the pure IL 352 (Fig. S2). Furthermore, it has been found that the GnP nanoadditives are placed along several 353 furrows on the worn surface. Therefore, mending effect takes place also resulting in a smoother 354 surface. From the mappings in Fig. 14, it can be concluded that the IL plays a more important 355 role for TTM/IL/h-BN than for TTM/IL/GnP, due to its stronger concentration in the tribofilm 356 of the worn surface. Taking into account Raman, SEM and roughness results it can be 357 concluded that for both nanodispersions the main tribological mechanisms are the formation of 358 the IL and nanoparticle tribofilms, as well as the mending effect due to nanoparticles. Thus, 359 positive synergies between the nanoadditives and the IL were found. 360 361 362 23 363 Fig. 14. Raman spectra and elemental map of the worn surface obtained with the 364 nanolubricant a) TTM + 2 wt% IL + 0.1 wt% h-BN and b) TTM + 2 wt% IL + 0.1 wt% GnP 365 366 4. Conclusions 367 In this work the following features were achieved: 368 1. Dispersions based on graphene nanoplatelets, GnPs, or nanoparticles of hexagonal 369 boron nitride, h-BN, with or without the IL tri(butyl) ethylphosphonium 370 diethylphosphate in an ester type base oil, triisotridecyltrimellitate (TTM), were 371 prepared. Three weeks after their preparation, none of the dispersions showed signs of 372 instability. 373 2. Rotational tribological tests were performed with a CSM standard tribometer under a 374 normal load of 20 N, the contact pair being AISI52100/AISI52100. The mean friction 375 coefficients obtained lubricating the contact with each one of the dispersions are lower 376 than the corresponding ones using TTM without additives. With respect to that obtained 377 with the neat oil, the maximum reduction of the friction coefficient is 33%, reached with 378 the TTM/IL/GnP nanodispersion. We should point out that there is no previous study of 379 GnP dispersions combined with both IL and base oil. 380 24 3. Wear was evaluated in terms of the width (WTW), the depth (WTD) and cross-section 381 area of the wear track, as well as the roughness of the worn surface. In comparison to 382 those obtained with the neat oil, the maximum reductions of mean WTW (44%), the 383 mean transversal area (66%) and the mean roughness (65%) correspond to the 384 TTM/IL/GnP nanodispersion whereas that of the mean WTD (32%) is obtained with 385 TTM/IL/h-BN. Tribofilm formation was confirmed by confocal Raman microscopy on 386 the worn surfaces. The results obtained for TTM/IL/h-BN are better than the only 387 previous study with dispersions Palm Olein TMP Ester/IL/h-BN at room temperature 388 [32]. 389 4. From the above results, it can be concluded that positive synergies between the IL and 390 h-BN or GnP as additives of TTM are found. 391 Declaration of competing interest 392 None. 393 Acknowledgments 394 It is a pleasure to thank Dr. Alfredo Amigo and Dr. María J. G. Guimarey (both from Applied 395 Physics Department, University of Santiago de Compostela) for kindly allowing to use a 396 refractometer and provide us unpublished data, respectively. Authors acknowledge Verkol and 397 Cytec Solvay Group for providing us the TTM and IL samples respectively. Authors would like 398 to thank the use of RIAIDT-USC analytical facilities, especially to Mr. Ezequiel Vázquez for 399 his useful advice. This work was supported by MINECO and the ERDF programme through 400 ENE2014-55489-C2-1-R and ENE2017-86425-C2-2-R projects, and by the Xunta de Galicia 401 (ED431E 2018/08, ED431D 2017/06 and GRC ED431C 2016/001). These funders also 402 financed the acquisition of the 3D Optical Profile (UNST15-DE-3156). 403 25 References 404 [1] K. Holmberg, A. Erdemir, Influence of tribology on global energy consumption, costs and 405 emissions, Friction 5 (2017) 263-284. 406 [2] S. Shahnazar, S. Bagheri, S.B.A. Hamid, Enhancing lubricant properties by nanoparticle 407 additives, Int. J. Hydrogen Energy 41 (2016) 3153-3170. 408 [3] M.I.H.C. Abdullah, M.F.B. Abdollah, N. Tamaldin, H. Amiruddin, N.R.M. Nuri, Effect of 409 hexagonal boron nitride nanoparticles as an additive on the extreme pressure properties of engine oil, 410 Ind. Lubr. Tribol. 68 (2016) 441-445. 411 [4] O.N. Çelik, N. Ay, Y. Göncü, Effect of nano hexagonal boron nitride lubricant additives on the 412 friction and wear properties of AISI 4140 steel, Particul. Sci. Technol. 31 (2013) 501-506. 413 [5] M.S. Charoo, M.F. Wani, Tribological properties of h-BN nanoparticles as lubricant additive on 414 cylinder liner and piston ring, Lubr. Sci. 29 (2017) 241-254. 415 [6] C.J. Reeves, P.L. Menezes, Evaluation of boron nitride particles on the tribological performance 416 of avocado and canola oil for energy conservation and sustainability, Int. J. Adv. Manuf. Technol. 89 417 (2017) 3475-3486. 418 [7] Q. Wan, Y. Jin, P. Sun, Y. Ding, Tribological behaviour of a lubricant oil containing boron nitride 419 nanoparticles, Procedia Eng. 102 (2015) 1038-1045 420 [8] G.M. Ay, Y. Göncü, N. Ay, Environmentally friendly material: Hexagonal boron nitride, Journal 421 of Boron 1 (2016) 66-73. 422 [9] F.U. Shah, S. Glavatskih, O.N. Antzutkin, Boron in Tribology: From Borates to Ionic Liquids, 423 Tribol. Lett. 51 (2013) 281-301. 424 [10] T.Q.P. Vuong, S. Liu, A. Van der Lee, R. Cuscó, L. Artús, T. Michel, P. Valvin, J.H. Edgar, G. 425 Cassabois, B. Gil, Isotope engineering of van der Waals interactions in hexagonal boron nitride, Nature 426 Materials 17 (2017) 152. 427 [11] J.M. Liñeira del Río, M.J.G. Guimarey, M.J.P. Comuñas, E.R. López, A. Amigo, J. Fernández, 428 Thermophysical and tribological properties of dispersions based on graphene and a 429 trimethylolpropane trioleate oil, J. Mol. Liquids 268 (2018) 854-866. 430 [12] S.S.N. Azman, N.W.M. Zulkifli, H. Masjuki, M. Gulzar, R. Zahid, Study of tribological properties 431 of lubricating oil blend added with graphene nanoplatelets, J. Mater. Res. 31 (2016) 1932-1938. 432 [13] W. Rashmi, M. Khalid, Y.L. Xiao, T.C.S.M. Gupta, G.Z. Arwin, Tribological studies on 433 graphene/tmp based nanolubricant, JESTEC 12 (2017) 365-373. 434 [14] D.W. Chang, J.-B. Baek, Eco-friendly synthesis of graphene nanoplatelets, J. Mater. Chem. A 4 435 (2016) 15281-15293. 436 [15] M. Zonggang, L. Weimin, Z. Shuxiang, Z. Feng, Functional Room-temperature Ionic Liquids as 437 Lubricants for an Aluminum-on-Steel System, Chemistry Letters 33 (2004) 524-525. 438 [16] Z. Mu, F. Zhou, S. Zhang, Y. Liang, W. Liu, Effect of the functional groups in ionic liquid 439 molecules on the friction and wear behavior of aluminum alloy in lubricated aluminum-on-steel 440 contact, Tribology International 38 (2005) 725-731. 441 [17] A.E. Jiménez, M.D. Bermúdez, P. Iglesias, F.J. Carrión, G. Martínez-Nicolás, 1-N-alkyl -3-442 methylimidazolium ionic liquids as neat lubricants and lubricant additives in steel–aluminium contacts, 443 Wear 260 (2006) 766-782. 444 [18] A.E. Jiménez, M.D. Bermúdez, F.J. Carrión, G. Martínez-Nicolás, Room temperature ionic 445 liquids as lubricant additives in steel–aluminium contacts: Influence of sliding velocity, normal load and 446 temperature, Wear 261 (2006) 347-359. 447 [19] J. Qu, J.J. Truhan, S. Dai, H. Luo, P.J. Blau, Ionic liquids with ammonium cations as lubricants or 448 additives, Tribology Letters 22 (2006) 207-214. 449 [20] A.-E. Jiménez, M.-D. Bermúdez, Ionic liquids as lubricants for steel–aluminum contacts at low 450 and elevated temperatures, Tribology Letters 26 (2007) 53-60. 451 [21] B.A. Kheireddin, W. Lu, I.C. Chen, M. Akbulut, Inorganic nanoparticle-based ionic liquid 452 lubricants, Wear 303 (2013) 185-190. 453