Moringa Oil and Carbon Phases of Different Shapes as Additives for Lubrication

Abstract

Vegetable oils in the lubricant field are largely studied. Their efficiency depends on their viscosity parameters and their fatty acid composition. The actions of moringa oil used as a lubricant base and as a lubricant additive have been shown in this work. Graphite, carbon nanofibers, and carbon nanodots are carbon phases of different shapes used as solid additives. The tribological performances of lubricant blends composed of between 0.5 and 1 wt.% of particles have been evaluated using a ball-on-plane tribometer under an ambient atmosphere. No additional surfactant was used. The positive and important actions of a small amount of moringa oil added in the lubricant formulas are demonstrated. The results obtained allow us to point out the influence of the type and shape of particles. Physicochemical investigations allow us to propose a synergistic effect between the particles and moringa oil as additives in dodecane.

1. Introduction

Vegetable oil in the lubrication field presents a real challenge. Researchers are attentively looking for ways to reduce fossil fuels and substitute them with viable alternatives. This trend is primarily due to the nontoxic and biodegradable characteristics of vegetable oils. Indeed, lubricating oils that are derived from mineral oil are not environmentally friendly and can induce threats to the ecosystem if not disposed of correctly. Natural oils developed from vegetable or animal precursors or biomass present better tribological properties than conventional industrial lubricants. Several vegetable oils with different fatty acid compositions (saturated and unsaturated fatty acids) and varied viscosities were investigated [1,2,3,4,5]. Reeves et al. deduced that natural oils presenting high oleic acid concentrations improve friction and wear performance by establishing densely packed monolayers on the lubricating surface [6]. Liu et al. show that the content of C18:1 fatty acid has a positive effect on the lubrication performances of vegetable oils, while the content of C18:3 compound has a negative effect, and the length of the carbon chain of fatty acids significantly affects their lubrication properties [7]. Moringa oil mainly constituted of oleic acid molecules is used in this study. Polar functional groups in the triacylglycerol molecule of vegetable oil orient themselves at the solid surface, building a close-packed monomolecular or multimolecular layer resulting in a surface film [8]. Different studies have shown that natural oils play a prevailing role in influencing the tribological performances of mineral/vegetable oil blends [6,9]. Chan et al. reviewed the physical and tribochemical properties of biolubricant base stocks and additives [10]. They agreed that the monounsaturated carbon chain as the oleic acid one is the best tradeoff in most base vegetable oil. The tribological additives for biolubricants are plant-derived compounds and polymers, particulates (CuO, TiO₂, and ZnO), and ionic liquids. They conclude that tuning the viscosity range of the lubricant and lubrication regime is suitable for the application requirements.

In this study, we compare the tribological properties of new natural nanometric particles obtained from seaweeds as additives to other carbon particles (graphite and carbon nanofibers). Carbon dots (CDs) are used as lubricant additives in water-based lubricants leading to better friction properties and a reduction in wear [11,12]. Guo et al. worked with nano-MoS₂ quantum dots (QDs) as additives in paraffin oil [13]. They showed that the lowest friction coefficient was 0.061 with 0.3 wt.% MoS₂ QDs added.

Spikes et al. studied different friction modifier additives in order to understand their operating mode [14]. In the case of nanoparticle mechanisms, the critical issue is to ensure that particles enter into the contact and adhere to metallic surfaces. Carbon-based nanoparticles have attracted attention due to their physical, chemical, and mechanical properties in field of tribology. These nanoparticles can be classified depending on their dimensionality: zero-dimensional (0D), such as fullerenes, one-dimensional (1D), such as carbon nanotubes, two-dimensional (2D), such as graphene, or three-dimensional (3D), such as nanodiamonds [15,16]. Lee et al. studied the influence of fullerenes on the friction performance of a mineral oil and showed that the optimal concentration was 0.5 wt.% [17]. Cornelio et al. found that the best tribological response for the rail steel in contact with wheel steel was obtained when multiwall carbon nanotubes (MWCNTs) were added at a concentration of 0.01% to base oil [18]. The mechanism is attributed to the formation of an amorphous carbon film that protected the surfaces. Nassef et al. showed reductions of up to 41% of the frictional behavior of mixtures prepared with 0.5, 1, 2, and 4 wt.% of graphene in rapeseed oil at room temperature [19].

The objective of this work is to compare the tribological performances in correlation with the dimensionality of carbon-based particles used as solid additives in both vegetable oil and mineral oil bases. Moringa oil performances were investigated as a base lubricant and as a liquid additive for lubrication. Graphite and carbon nanofibers have been studied in previous studies with similar conditions [20]. These results are discussed and faced with those obtained for carbon nanodots synthetized from Sargassum seaweeds. The influence of weak amounts of vegetables and the optimal concentration were evaluated for the lubricant formulas depending on the dimensional shapes of the particles.

2. Materials and Methods

2.1. Materials

Moringa oil (MO) used was extracted by Phytobokaz laboratory Guadeloupe and is mainly composed of monounsaturated fatty acid (

Table 1. Composition of moringa oil.

Fatty Acid Methyl Ester		% Mole Fraction
Palmitic	C16:0	6.09
Palmitoleic	C16:1	1.94
Stearic	C18:0	3.77
Oleic	C18:1	75.33
linoleic	C18:2	0.90
Linolenic	C18:3	0.29
Arachidic	C20:0	2.47
Behenic	C22:0	5.67
Lignoceric	C24:0	1.01

). Dodecane, ReagentPlus, 99%, used in this study as the base oil, was provided by Sigma Aldrich (St. Louis, MO, USA). Figure 1 presents SEM micrographs of the carbon materials used in this study as solid friction reducers: pristine carbon nanofibers (CNFs), and graphite and nanocarbon dots (CDs). CNFs with high purity (>90%) presented lengths of about 2–20 µm and diameters of 150 ± 3 nm, and were supplied by MER industry (Or Yehuda, Israel). Graphite particles came from Timcal Society (Bodio, Switzerland) with a thickness about 100 nm and an average size of 40 µm. CDs were synthetized in our laboratory from Sargassum seaweeds collected along the coast of Guadeloupe [21]. The maximum size distribution of the CDs was 10 nm.

All mixtures were prepared with simple weighing with a precision of 0.01 mg. Blends with MO as liquid additive contained 0.5, 1, 1.5, 2, and 3 wt.% in dodecane. Lubricant formulas with carbon particles as solid additives were prepared by adding 0.5 and 1 wt.% in MO and dodecane base oil. The last lubricant formulas were realized with MO and solid particles as additives in dodecane with the same weighing technique. MO is added at 1 and 2 wt.%. The particle’s weight proportions have been chosen according to precedent results obtained with blends containing dodecane base oil, either 1 wt.% of graphite, 0.5 wt.% of CNFs, or 0.5 wt.% for CDS. The dispersion of the different blends was obtained in an ultrasonic bath over 5 min.

2.2. Tribological Tests

The tribological properties were evaluated at room temperature (25 °C) using a reciprocating ball-on-plane tribometer (built in GTSI laboratory). A steel ball (AISI 52100) with a sliding speed of 4 mm.s⁻¹ at a frequency of 1 Hz rubbed against a steel plane (AISI 52100).

The tested material was deposited on the plane, and a normal load F_N of 10 N was applied (Figure 2). According to Hertz’s theory, the contact diameter is 140 µm, with a maximum contact pressure of 1 GPa. The friction experiments were about 2000 cycles. These conditions were chosen in order to perform the tribological tests in boundary lubrication. A cycle corresponds to an alternative motion of the ball. For all experiments, steel balls were used with initial roughness (50 µm), while planes were polished with abrasive disks (roughness of about 350 µm) to generate multidirectional stripes in order to favor the presence of solid particles in the sliding contact. Before friction experiments, balls and planes were successively cleaned in ultrasonic baths of acetone and ethanol. The tangential force F_T was measured with a computer-based data acquisition system. The friction coefficient was defined as $\mu ={\displaystyle \raisebox{1ex}{$F_T$}\!\left/ \!\raisebox{-1ex}{$F_N$}\right.}$. About 10 friction tests were repeated for each tested compound. The solid materials (about 0.5 mg) were deposited on the plane’s surface. A drop (about 20 µL) of the selected mixture was deposed on the plane’s surface before the friction test. The blends with MO and dodecane are referred to as liquid conditions, and the friction coefficient measured is noted as ${\mu }_{wt.\% MO+dodecane}$. In the case of mixtures containing solid particles, the notations are ${\mu }_{wt.\% particles+dodecane}$ or ${\mu }_{wt.\% particles+MO}$ and ${\mu }_{wt.\% MO/dodecane+wt.\% particles}$. The antiwear properties of the compounds are evaluated by comparing the contact diameter measured on the ball to the theorical one (140 µm in equivalent conditions).

2.3. Physicochemical Characterizations

In order to identify the functional groups in the MO/dodecane blends, Fourier transform infrared spectroscopy (FTIR) was performed with the uncertainty being approximately ±3 cm⁻¹. A PerkinElmer Spectrum Two spectrometer with a range from 4000 to 500 cm⁻¹ wavenumbers and a resolution of 4 cm⁻¹ was used. For these blends, viscosity parameters were measured using a Modular Compact Rheometer (Anton Paar, Graz, Austria). The share rate was from 0.01 to 1000 s⁻¹, and the contact cone/planes at ambient temperature were the conditions. The cone diameter was 50 nm with an angle of 2°, and the plane diameter was 50 nm. The particles and their corresponding tribofilms were characterized using Scanning Electron Microscopy (SEM) using secondary electron imaging (FEI Quanta 250 microscope, FEI Company, Hillsborough, CA, USA), and using Raman spectroscopy. This latter method was performed with an HR 800 Horiba multi-channel spectrometer (HORIBA France SAS, Longjumeau, France) using a Peltier-cooled charged coupled device (CCD) detector for signal recording. A green laser exciting light (532 nm) was used to record the Raman spectra. The Raman spectrometer (HORIBA France SAS, Longjumeau, France) ran with a notch filter pre-monochromator and a 300 lines/mm holographic grating monochromator. Our experimental conditions led to a probe diameter of 5 µm and a wavenumber resolution of 1.5 cm⁻¹. In order to prevent the carbon materials from irradiation damage, the laser power was 30 mW with an acquisition time in the 10 to 60 s range.

All lubricants and blends prepared and the methods used are indicated in

Table 2. All lubricants prepared and experimental methods used in this study.

Lubricant Base	Solid Additives	Liquid Additive	Experimental Methods
Dodecane	0.5 and 1 wt.% of graphite— 0.5 and 1 wt.% of carbon nanofibers— 0.5 and 1 wt.% of carbon nanodots		Ball-on-plane tribometer
Moringa oil	0.5 and 1 wt.% of graphite— 0.5 and 1 wt.% of carbon nanofibers— 0.5 and 1 wt.% of carbon nanodots		Ball-on-plane tribometer
Dodecane		0.5, 1, 1.5, 2, 3 wt.% of moringa oil	Ball-on-plane tribometer Viscometer FTIR
Dodecane	1 wt.% of graphite	1 and 2 wt.% of MO	Ball-on-plane tribometer Raman spectrometer SEM
Dodecane	0.5 wt.% of carbon nanofibers	1 and 2 wt.% of MO	Ball-on-plane tribometer Raman spectrometer SEM
Dodecane	0.5 wt.% of carbon nanodots	1 and 2 wt.% of MO	Ball-on-plane tribometer Raman spectrometer SEM

.

3. Results

3.1. Moringa Oil as Additive in Dodecane

In previous works, we have shown that the friction properties of dodecane are significatively improved in the presence of MO [20,22]. Figure 3a presents tribological results obtained for pure MO, pure dodecane, and blends. A small amount of MO in the blends led to an improvement of about 50% in the friction performances of dodecane. Considering pure oils, the friction properties of pure MO are better than pure dodecane ones, as ${\mu }_{pure dodecane}=0.18\pm 0.02$, whereas ${\mu }_{pure MO}=0.070\pm 0.005$. The antiwear properties of pure MO are also better than those of pure dodecane, as ${\mathrm{\varnothing }}_{pure MO}=150\pm 10 \mathsf{\mu }\mathrm{m}$ and ${\mathrm{\varnothing }}_{pure dodecane}=280\pm 10 \mathsf{\mu }\mathrm{m}$, confirming the excellent properties of vegetable oils. In the case of MO/dodecane blends, an important reduction in friction and wear is observed. A progressive diminution in the friction coefficient values and then a stabilization is observed between 1 and 2 wt.% of MO in dodecane. In the case of the addition of 3 wt.% of MO in dodecane, the friction coefficient value was higher. Starting from ${\mu }_{0.5wt.\%MO+dodecane}=0.13\pm 0.010$, the friction coefficient decreases down to $0.1\pm 0.005$ for blends composed with 1.5 wt.% to 2 wt.% of MO and increases up to ${\mu }_{3wt.\%MO+dodecane}=0.120\pm 0.005$. The contact diameter is reduced from $180$ to $145\pm 5 \mathsf{\mu }\mathrm{m}$ as a function of the wt.% of MO added. The addition of MO in dodecane leads to an improvement in the antiwear properties of the base. If MO is added in concentrations above 1 wt.%, the contact diameters are close to that of pure MO. The beneficial role of MO in the antiwear properties of the base is obtained with a small wt.% addition (1.5 wt.%).

We have also shown that the presence of MO in the blends is not detectable using FTIR. Figure 3b presents the FTIR analyses that allowed us to identify the characteristics of groups of MO in the blends. Oleic fatty acid molecules characterized by C-H bending, C=O stretching, and C=C corresponding to peaks in the 1400 and 1800 cm⁻¹ area are weak in pure MO. Consequently, the presence of oleic acid is hardly detectable in the different MO/dodecane blends. In order to determine if the friction reduction improvement is influenced by the viscosity of the lubricant base, we compared the viscosity of pure MO and pure dodecane to those of the blends. Pure moringa oil’s viscosity parameter is higher than dodecane: ${\nu }_{pure MO}=87 \mathrm{m}\mathrm{P}\mathrm{a}\mathrm{s}$ while ${\nu }_{pure dodecane}=1.38 \mathrm{m}\mathrm{P}\mathrm{a}\mathrm{s}$. But whatever the percentage of MO added in dodecane was, the blends presented similar viscosity: ${\nu }_{1wt.\%MO+dodecane}\approx {\nu }_{2wt.\%MO+dodecane}\approx {\nu }_{3wt.\%MO+dodecane}\approx 2.37 \mathrm{m}\mathrm{P}\mathrm{a}\mathrm{s}$. The addition of a small amount of MO improved the friction properties of dodecane, but whatever the percentage of MO added was, the viscosity was not modified. According to the friction results obtained for these blends, base formulations of 1 and 2 wt.% of MO in dodecane were selected.

3.2. Solid Particles as Additives

Graphite particles are well known to present excellent tribological properties in air atmosphere as ${\mu }_{pure graphite}=0.09\pm 0.005$. The friction coefficient of carbon nanofibers (CNFs) is ${\mu }_{pure CNFs}=0.14\pm 0.01$ in our experimental conditions. Carbon nanodots (CDs) synthesized from sargassum are studied as new solid particles for lubrication. The friction coefficient under air atmosphere is ${\mu }_{pure CDs}=0.25\pm 0.01$. Friction properties of the blends composed of between 0.5 and 1 wt.% of particles are evaluated.

3.2.1. In Dodecane Base

Figure 4a presents the friction coefficient values obtained for the blends with a dodecane base. The addition of weak amounts of graphite and CNF particles improves the tribological properties of dodecane. For both particles, a reduction of about 70% of µ was obtained, but an influence of the particle’s shape was observed. Indeed, ${\mu }_{0.5wt.\% graphite+dodecane}=0.09\pm 0.005$ while ${\mu }_{1wt.\% graphite+dodecane}=0.06\pm 0.005$. The opposite behavior is observed for CNF additives. The best value is obtained with 0.5 wt.% of CNFs; ${\mu }_{0.5wt.\% CNFs+dodecane}=0.065\pm 0.010$ while ${\mu }_{1wt.\% CNFs+dodecane}=0.1\pm 0.01$. In the case of CD additives, the friction coefficient values were higher than the pure dodecane coefficient: ${\mu }_{0.5wt.\% CDs+dodecane}=0.2\pm 0.01$ and ${\mu }_{1wt.\% CDs+dodecane}=0.23\pm 0.01$. These results suggest the significant influence of the percentage and the shape of particles on the tribological properties of dodecane. According to these results, the best amounts for each particle are 1 wt.% of graphite and 0.5 wt.% of CNFs and CDs, in this study.

3.2.2. In MO Base

Figure 4b presents the friction coefficient values obtained when solid particles are added to the MO base. No evolution of the friction coefficient is observed whatever the type of particles were. All of the coefficient values are higher than the pure MO ones. In the case of CNF and CD additions, the coefficients are similar for 0.5 and 1 wt.% of solid particles. For graphite, a weak difference is noted in favor of the 1 wt.% blend, ${\mu }_{0.5wt.\% graphite+MO}=0.1\pm 0.05$, while ${\mu }_{1wt.\% graphite+\mathrm{M}\mathrm{O}}=0.085\pm 0.005$. The influence of the viscosity parameter is shown.

3.2.3. In 1 wt.% of MO/Dodecane Mixtures

Figure 5a presents the friction curves obtained for particles after addition in 1 wt.% of MO in a dodecane base. The percentage of particles added in the mixture was selected after considering the precedent results. The friction behavior is different according to the type of particles. In the case of graphite, the coefficient decreases down to 0.06 after 500 cycles, and then increases up to 0.07 until the end of the experiment. For CNFs, an induction period about 800 cycles is observed followed by a stabilization of µ at 0.06 until the end. The best coefficient is obtained for CD mixtures. The initial friction coefficient value is weak (0.08) and decreases down to 0.055 after ten cycles. On Figure 6a, we can see that the induction period is reduced for CDs due to their small particle size. However, the friction coefficient increases after 1000 cycles up to 0.11 ± 0.003. At 2000 cycles, the friction coefficient value of this mixture is higher than the 1 wt.% of MO/dodecane blend.

3.2.4. In 2 wt.% of MO/Dodecane Mixtures

Figure 5b presents the friction curves obtained in the case of particles added in 2 wt.% of MO in dodecane base. The coefficient values are more important for mixtures containing CNFs and CDS. Indeed, in the case of CNFs, after a weak induction period, the coefficient decreased down to 0.08 and then increases up to 0.1 until the end of the experiment. This value corresponded to the friction of the 2 wt.% MO/dodecane pure base. For the CD mixture, the starting coefficient is 0.1 and then 0.09 until the end. The best result is obtained for the graphite mixture. After 500 cycles, the friction coefficient decreases down to 0.05 ± 0.05 until the end. No influence of the particle’s size on the induction period was observed.

3.3. Physicochemical Characterization

No significant difference in the contact diameter measured on the ball after these last friction experiments has been noted. The values are about 140 µm, corresponding to Hertz’s first theory.

3.3.1. CNFs

SEM and Raman spectroscopy investigations were carried out at the end of the friction test for 1 wt.% MO/dodecane + 0.5 wt.% CNFs lubricant. The results are presented in Figure 6a,b. Figure 6a shows that the tribofilm is homogenous and thick, as the initial stripes of the steel plane are not visible. Raman spectra recorded on the middle and outside of the wear scar are different (Figure 6b). The two major carbon characteristic peaks corresponding to D mode near 1350 cm⁻¹ are associated with disorder, and another one, the G + D’ mode near 1600 cm⁻¹, was detected [23]. The G band is attributed to the E_2g vibration mode of the graphitic lattice. The G and D’ Raman peaks reflect the evolution of the number of graphene layers. Indeed, an enlargement of both Raman bands is observed, indicating a modification of the structure of the CNFs and high mechanical solicitations on CNFs particles during friction experiments.

3.3.2. Graphite

The best results were obtained when 1 wt.% of graphite was added to the 2 wt.% of MO/dodecane base. Raman spectrometry analyses were performed on friction scars obtained on the steel plane at the end of the friction (Figure 6d). No differences between the Raman spectra recorded on particles outside the scar trace and on the tribofilm were visible, suggesting that the friction process does not lead to a significant structural evolution in graphite particles (Figure 6c). The tribofilm built during the friction test shows antiwear properties, as the initial stripes of the steel plane are visible under the tribofilm (Figure 6d). Moreover, parts of the tribofilm are extracted during friction test, characterizing the weak adhesion of the steel plane.

3.3.3. CDs

In the case of the best CD mixtures, the friction coefficient value increases after 1000 cycles. To evaluate the presence of CDs in the contact during the friction test, in situ Raman analyses were performed (Figure 7). The presence of CD particles is detected on the Raman spectra using fluorescence. Figure 7a presents a Raman spectrometry analysis recorded at the beginning and at the end of the friction test. Fluorescence evaluated using the area calculated under Raman spectra allows us to identify the presence of CDs in the contact. Figure 7b shows the evolution of the friction coefficient superimposed with the calculated area of each Raman spectra recorded during the friction experiment. We noted a decrease in the area during the friction experiment, characterizing an absence of CDs at the end. The particles appeared to come out of the sliding contact during the friction test.

4. Discussion

The influence of the lubricant base on the tribological properties with different carbon particles was evaluated. In previous papers, we have shown the viscosity effect on the tribological properties of mixtures leading to a reduction in the friction coefficient [24]. These results are confirmed with this study at the addition of each particle. Indeed, the viscosity of MO is higher than dodecane one, and no friction reduction is obtained for the blends composed of the MO base. The addition of solid particles leads to an increase in the friction coefficient values compared to the pure MO one. The blends with 1 wt.% of graphite in the MO base present a better result. However, in the case of blends composed of a dodecane base, for an equivalent percentage of particles added, an important reduction is observed for graphite and CNF particles. The same friction coefficient value is obtained for two different percentage of particles: ${\mu }_{1 wt.\% graphite+dodecane}={\mu }_{0.5 wt.\% CNFs+dodecane}\approx 0.06$. For blends containing CD particles in a dodecane base, no reduction is observed. Whatever the percentage of particles is, the friction coefficient value is similar and higher than the pure dodecane one. We can suppose that the number of CD particles is not adequate to influence the tribological properties. These results provide an influence on the size and shape of the particles on the tribological properties of these lubricant blends. In this study, we have shown that blends containing carbon particles with high size have a high optimal concentration to assure an important friction reduction and vice versa. A connection between the viscosity parameters and the optimal particle concentration should be investigated. In the literature, the tribological properties of vegetable oils as a lubricant base is extensively studied [2,25]. Nasr et al. studied the effect of adding different carbon-based nanoparticles in mineral oils and vegetable oils [26]. Our results are in opposition. Indeed, they found friction coefficient values to be higher in mineral oil than in rapeseed oil, with 0.1, 0.5, 1, and 2 wt.% of particles. They attributed the mechanism of action of the particles to the formation of an amorphous film that protected the surfaces by an agglomeration of the carbon-based nanoparticles due to the van der Waals forces existing between each particle. They suggested that the nanoparticles’ influence on the flow patter of the lubricant molecules reduced the internal friction. However, in recent studies, the friction coefficient values obtained for mixtures composed of rapeseed and graphene are in accordance with our results [19].

The beneficial effect of MO as a liquid additive has been highlighted in the presence of solid particles or not for mixtures composed of dodecane bases. Indeed, the addition of a weak percentage of MO in dodecane leads to an improvement in the tribological properties of dodecane. This is due to the presence of fatty acid molecules. In the literature, vegetable oils and their blends with mineral oils have been largely investigated; however, they worked with important proportions (10, 20 … 50%) [27,28,29]. To explain the reduction, researchers suppose an effect of the thickness film of fatty acid molecules formed on the steel surfaces. Fry et al. have demonstrated that the properties of organic friction with adsorbed layers govern the friction by forming an adsorbed layer with critical thickness necessary to provide low friction [30]. Studies have shown that friction coefficient and wear rate are dependent on the adsorption energy of the lubricant by physisorption or symmetric/asymmetric chemisorption [31,32]. In this study, we found the optimal concentration of the MO liquid additive in dodecane. With 1 and 2 wt.% of MO, the friction reduction is stable. With 3 wt.% of MO, this increases up, leading to conclusion at the given saturation. The evolution of the friction coefficient values according to the percentage of MO added in dodecane points to an effect of the adsorbed film thickness of fatty acid molecules on the tribological performances of dodecane.

In the presence of solid additives, MO as a liquid additive has an important effect on the tribological properties of the blends. Indeed, these lubricant formulas present different friction behavior according to the amount of MO added. In the presence of 1 wt.% of MO, the best results have been obtained with 0.5 wt.% of CDs. The starting coefficient is about 0.055 during 1000 cycles, and afterward increases up to 0.12. In the case of the mixtures composed of graphite and CNF particles, the friction coefficients decrease and remain stable until the end of the experiment at a more important value. Unfortunately, CD particles seem to come out of the sliding contact, probably due to surface properties of the particles. Graphite and CNFs present hydrophobic surfaces, whereas CDs present hydrophilic ones [33]. These results allow us to conclude the presence of an action of adsorption of fatty molecules on the particle’s surface. At the addition of 2 wt.% of MO, the lubricant formula containing 1 wt.% of graphite presents a more important friction reduction.

The results of this study confirm a positive action of adsorption of fatty acid molecules associated with the influence of the presence of liquid in the sliding contact to explain the friction improvement. We are able to propose a nanofluid mechanism in accordance with the type of particles (Figure 8). A protective film is formed on the steel surface facilitating the sliding, but mainly the effect is due to friction nanofluid mechanism. In the case of CNFs, particles are arbitrarily oriented, explaining the modification of the particle’s structure and the high friction coefficient of the lubricant mixture. Bhaumik et al. showed a rolling effect mechanism for SWCNTs [34]. The graphite and CDs particles orientated in the sliding direction. Particles and liquid have to be present in the sliding contact to reduce the coefficient. Moreover, the presence of weak molecules of fatty acids has an important action in this process of the nanofluid mechanism and in the adsorption on the particles’ surfaces.

5. Conclusions

Carbon phases are solid particles generally used as friction-reducer additives for lubricants. Different new lubricants elaborated with a vegetable oil and with moringa oil as an additive or as a lubricant base were investigated. The main quantitative results are as follows:

• Moringa oil as a lubricant base is not efficient due to its high viscosity parameters. Indeed, with a similar composition, an important friction reduction is obtained for blends composed with a dodecane base, contrary to moringa base oil.
• Moringa oil used as a liquid additive for lubricants has a beneficial action on the friction of mineral base oil and on the tribological properties of blends according to the shape of the carbon-phase when used as additives.

The comparison between the tribological performances of dimensionality of carbon-base particles as solid additives show an influence of the type, the shape, and the amount of the solid carbon particles as additives on the friction properties of lubricant mixtures. Indeed:

• The 0D particles could be an excellent candidate for the biolubrication field, but more investigations will have to be realized to increase their stability during friction experiments. The CDs used are natural carbon phases, and present good results in the blends.
• The 1D particles are additives and are less interesting for lubricant blends due to the random movements in the nanofluid during the sliding.
• The 2D particles present the best stable friction properties. However, the additive composition of the lubricant blends composed of graphite is higher.