Tribology Letters

, Volume 47, Issue 1, pp 91–102 | Cite as

Tribological Effects of BN and MoS2 Nanoparticles Added to Polyalphaolefin Oil in Piston Skirt/Cylinder Liner Tests

  • Nicholaos G. Demas
  • Elena V. Timofeeva
  • Jules L. Routbort
  • George R. Fenske
Original Paper

Abstract

We report in this article the friction and wear results of polyalphaolefin (PAO 10) base oil with the addition of 3 wt% boron nitride (BN) and molybdenum disulfide (MoS2) nanoparticles with nominal size of 70 and 50 nm, respectively. The formulations were tested using cast iron cylinder liner segments reciprocating against aluminum alloy piston skirt segments at 20, 40, and 100 °C. The results showed that, at a load of 250 N and a reciprocating frequency of 2 Hz, BN did not lower friction whereas MoS2 nanoparticles were very effective at reducing both friction and wear, compared with the base oil. The viscosities of both formulations were similar to the base oil, which allowed for a direct comparison between them. Raman spectroscopy showed the formation of an aligned MoS2 layer on the cast iron liner surface, which most likely functions as a tribofilm. In the case of the cast iron liner tested with BN nanolubricant, no traces of BN were found. The effect of surfactants was also studied, and it was found that some surfactants were not only beneficial in dispersing the nanoparticles in oil, but also in producing some reduction in friction and wear, even when used as stand-alone additives in PAO 10.

Keywords

Nanolubricants Nanoparticles Surfactants Additives Cast iron Cylinder liner Piston skirt Boron nitride (BN) Molybdenum disulfide (MoS2Polyalphaolefin (PAO) 

1 Introduction

In the internal combustion engine, the piston skirt’s contribution to the total mechanical friction losses of the piston/cylinder system is significant and comparable to that of piston rings, which accounts for more than 40 % of the total mechanical engine losses [1, 2, 3]. Researchers have realized that typical materials and lubricants used today as well as a variety of traditional surface treatments may not provide sufficient protection to meet the requirements of reliability, performance, oil consumption, and emissions in the internal combustion engine [4, 5]. Depending on the operating conditions, reduction in friction and wear can be achieved through optimization of the surface topography of the piston rings or cylinder liner, coating the piston and piston rings with low friction films, or through the introduction of additive packages to lubricant formulations [4, 5].

Nanolubricants (dispersion of nano-sized materials in oils) are emerging concepts in lubrication, some of which have demonstrated reduction in friction and wear of moving parts [6, 7, 8, 9, 10, 11, 12]. Nanomaterial additives may help in counteracting friction, saving energy, increasing the efficiency of dynamic transmission, dampening vibrations, and reducing noise [5]. While well-known solid lubricants, such as graphite, h-BN, and transition metal dichalcogenides, owe their lubricity to a unique layered structure, solid nanoparticles can engage additional lubrication mechanisms. Tribochemical film formation is often encountered, and even nanoscale ball bearing mechanisms have been proposed [6, 7, 8]. Furthermore, nanoparticles, due to their small size, may enter the contact area, which can have a positive effect. The main challenge with nanoparticles being used as additives in lubricants is their stability. Compatibility with the lubricating oil can be improved with surface modification techniques using organic compounds. Chalcogenides have been successfully prepared using a surface modification agent [13]. Even though a great effort has been put forth by researchers, there are still difficulties associated with the synthesis of many nanoparticles and their stability in suspensions.

There are reports in the literature demonstrating tribological improvements due to the addition of nanoparticles, but no clear mechanisms for the improvements have been proposed. Evidences showing tribological benefits have emerged from laboratory experiments. Actual engine tests tend to be expensive and time consuming, and are sparse in the open literature [14]. In this study, segments from a commercial piston/cylinder system were tribologically tested using a test rig that utilizes reciprocating motion and offers a fair compromise between actual and simple laboratory tribological tests. An investigation of surfactant ability to stably suspend BN and MoS2 nanoparticles was conducted, and the surfactant’s contribution to viscosity and tribological properties was studied. The cast iron samples subjected to friction tests at three temperatures were studied using Raman spectroscopy to identify chemical changes on the surface. Formation of tribofilms observed using Raman spectroscopy was correlated to the reduced friction and wear for some nanolubricant formulations.

2 Experimental Procedure

2.1 Samples and Equipment

The specimens used in this study were extracted from commercial heavy-duty diesel engine components. During all machining operations, the original surfaces of both piston and liner were protected to retain the original surface roughness and pattern. The skirt specimens, made of an aluminum alloy (Table 1), were 19 × 19 × 6.35 mm, and the gray cast iron liner segments were 50 × 38 × 8.5 mm. Circumferential grooves were present on the surface of the skirt specimens from the original manufacturing of the piston. The liner was plateau honed. The surface parameters of the liner are shown in Table 2. Figure 1a, b shows a photograph of the skirt segment, and a micrograph of the liner surface, respectively. A pronounced pattern of grooves can be seen on the skirt, while machining marks (cross-hatching) on the surface of the liner are evident. A photograph of the samples installed in the test rig is shown in Fig. 2. The cylinder liner was mounted onto a reciprocating table at the bottom of the test rig, while the piston skirt was stationary. The back surface of the skirt specimens was machined flat with a spherical radius groove, so that when the ball-ended holder fits into the groove, it gets self-aligned during testing, ensuring a proper conformal contact. A small amount of oil (0.3 ml) was applied at the interface of the samples at the start of each test. The tests were conducted with 2 Hz reciprocating frequency. Heating elements were embedded into the reciprocating table, and the temperature was controlled by a temperature control unit. A normal load of 250 N was applied with a pneumatic spring and measured with a force transducer, while the friction force was measured using another force transducer. Each test was performed for 3 h to get a representative evolution of friction over time. The same skirt and liner segments were used for each formulation at the three different temperatures (20, 40, and 100 °C), and the tests were repeated at least twice. After break-in, no significant variability was observed.
Table 1

Aluminum alloy piston composition

Element

Al

Si

Ti

Fe

Ni

Cu

Zr

At.%

76.2 ± 0.33

16.16 ± 0.12

0.11 ± 0.03

0.36 ± 0.02

2.25 ± 0.03

4.39 ± 0.04

0.11

Table 2

Topographical parameters of plateau-honed liner used in this study

Parameter

Value

Sa (roughness average)

0.62 μm

Sq (root mean square)

0.84 μm

Ssk (surface skewness)

−0.46

Sku (surface kurtosis)

17.4

Sds(density of summits)

0.0271/μm2

Spk (reduced peak height)

2.28 μm

Sk (core roughness depth)

1.56 μm

Svk (reduced valley depth)

1.50 μm

Fig. 1

Microscope images of a skirt segment b original cast iron surface

Fig. 2

Photographs of samples used in this study

The wear tracks on the liners were examined optically using an Olympus STM6 microscope after the tests were run.

Scanning electron microscope (SEM) images of the nanopowders used in this study were taken using Hitachi S-4700.

The dynamic viscosities of nanolubricants oil formulations at shear stress between 0.5 and 3.5 N/m2 and temperatures between 15 and 100 °C were measured using a Brookfield DV-II + rotational type viscometer (Brookfield Engineering Laboratories, Inc.) with the SC4-18 spindle (instrument error ~2 %).

The Raman microscope (Renishaw, the UK) with laser excitation of 633 nm was used at room temperature to examine the nanolubricants and the liner surfaces.

3 Results and Discussion

3.1 Suspension Stability and Viscosity of Nanolubricants

Nanolubricants were prepared from PAO 10 basestock oil with 3 wt% nanoparticles. Boron nitride (BN) and molybdenum disulfide (MoS2) nanoparticles with nominal sizes of 70 and 50 nm were obtained from Lower Friction, a division of M.K IMPEX Corp. The manufacturer’s specifications of the particle sizes were used in the selection of materials for this study. Dynamic light scattering measurements in diluted suspensions confirmed average particle sizes of 70 ± 10 nm for BN particles. Samples of MoS2 showed a bimodal distribution of particle sizes with the main fraction at 50 ± 10 nm. The samples for SEM were prepared from diluted suspension of nanoparticles in ethanol by placing a droplet of suspension onto a silicon wafer. Fast evaporation of the solvent provides good representation of particle size distribution (Fig. 3). BN nanoparticles were fairly spherical with individual particle sizes ranging between 50 and 300 nm, while MoS2 powder had a significantly wider range of sizes between 50 and 2,000 nm. Various surfactants were tested for dispersing the nanoparticles in PAO 10 oil. The results of all the tests were compared to those of pure PAO 10 baseline; nanolubricants with and without the surfactant were compared as well.
Fig. 3

SEM images of a h-BN (nominal size 70 nm) and b MoS2 (nominal size 50 nm)

In the absence of surfactants, nanoparticles get separated out of suspension within an hour. Successful dispersion of inorganic nanoparticles in non-polar PAO 10 oil required a study of suspension stability. Five different surfactants were tested: sodium dodecyl sulfate (#1), benzethonium chloride (#2), benzalkonium chloride (#3), Triton™ X102 (#4), and oleic acid (#5). Their chemical formulae and functionalities are given in Table 3.
Table 3

Surfactants used in this study

Surfactant

Functionality

#1 Sodium dodecyl sulfate Open image in new window

Anionic

#2 Benzethonium chloride Open image in new window

Cationic

#3 Benzalkonium chloride Open image in new window

Cationic

#4 Triton™ X 102 Open image in new window

Non-ionic

#5 Oleic acid Open image in new window

Anionic

Nanolubricants containing 1 wt% of surfactant and 3 wt% of either BN or MoS2 nanoparticles were sonicated using a Branson Sonifier, S450 horn. Surfactants #1, #2, and #5 produced stable suspensions of BN in PAO 10 (inset to Fig. 4a). Viscosity of stable suspensions was measured before friction and wear tests to investigate the shear and temperature dependences of suspensions and identify the formulation having viscosity close to that of PAO 10 base fluid. In BN suspensions, increase of viscosity depended on the type of surfactant as seen in Fig. 4a. Benzethonium chloride (S#2) provided the lowest viscosity, indicative of good nanoparticle dispersion. MoS2 suspensions were less sensitive to the nature of surfactant, all showing small increase of viscosity (Fig. 4b). All suspensions showed shear rate dependence, especially at low temperatures (inset in Fig. 4b). For simplicity, the viscosities of various fluids were compared at a shear stress of 2.5 N/m2 (Fig. 4b). As the temperature increased, the viscosity values and shear dependence decreased, approaching the corresponding values of the pure PAO 10. Formulations with S#2 and 3 wt% of both BN and MoS2 nanoparticles were chosen for friction and wear tests since they showed similar viscosities (Fig. 4b), which is important for the interpretation of friction tests and for defining the lubrication mechanisms.
Fig. 4

a Viscosities of 3 wt% BN suspensions in PAO 10 for a fixed shear stress of 2.5 N/m2. Inset in a illustrates visual appearance of sonicated 3 wt% BN suspension with 1 wt% of various surfactants after 1 day at rest. b Viscosities of nanolubricants at various temperatures at 2.5 N/m2 shear stress. Inset in b illustrates typical, for the studied nanolubricants, shear-dependent behaviors, especially at low temperatures

3.2 Friction

The coefficient of friction as a function of time was measured at 20, 40, and 100 °C for the selected lubricant formulations: PAO 10, PAO 10 with S#2, PAO 10 with 3 wt% BN—with and without S#2, and PAO 10 with 3 wt% MoS2—with and without S#2 (Fig. 5). It should be noted that the data at 100 °C are more representative of actual operating conditions than those at 20 and 40 °C.
Fig. 5

Graphs showing coefficient of friction as a function of time for PAO 10 and PAO 10 containing 3 wt% BN or 3 wt% MoS2, with and without surfactant #2 at 20, 40, and 100 °C

A continuous decrease in coefficient of friction was observed over time during the tests with pure PAO 10 at 20 °C (Fig. 5a). Morphological changes (wearing off the topmost features) in the surface are responsible for this behavior. The same trend in friction behavior was observed in PAO 10 with surfactant (Fig. 5b); however, the values of coefficient of friction were lower than those in pure PAO 10 throughout the test duration. It is suggested that interactions between the surfactant and the sample cause formation of a tribochemical film that, in addition to the slight increase of viscosity of the formulation due to the presence of surfactant, contributes to the reduction in the coefficient of friction. The trend was observed both at 20 and 40 °C. Furthermore, as will be discussed in the next section, little wear is measured on the skirt specimen at these temperatures which would further support the formation of a tribochemical film.

Introduction of BN nanoparticles (Fig. 5c) did not change the values of coefficient of friction significantly, but eliminated the gradual decrease of coefficient of friction observed in pure PAO 10 (Fig. 5a). This result might be attributed to the physical mechanisms of nanoparticles filling out the liner voids and valleys, thus providing a hydrodynamic effect and preventing morphological changes responsible for a decrease in coefficient of friction similar to that observed during the tests with pure PAO 10. Addition of the surfactant to the PAO 10 + BN mixture resulted in better stability of the suspension, as well as change in the friction behavior. However, it is clearly seen from Fig. 5d that the effect of the surfactant on friction is stronger than that of BN nanoparticles.

The coefficient of friction increased with temperature in the majority of the tested lubricant formulations (Fig. 5a–d). This is an expected result, because of a decrease of viscosity with temperature, leading to a shift from the mixed lubrication regime toward the boundary regime, where the increased asperity interaction results in s higher coefficient of friction. An exception to this temperature trend was observed for the formulation with MoS2 (Fig. 5e–f). At 20 and 40 °C, the frictions in PAO 10 + MoS2 were very similar to those in PAO 10 + BN formulations. A significant decrease in the coefficient of friction was observed in PAO 10 + MoS2 formulation as the temperature was increased to 100 °C. At this temperature, the contact is in the boundary lubrication regime. In this regime, as was observed in the previous cases, the friction was high. However, the presence of MoS2 resulted in a significant reduction in coefficient of friction (Fig. 5e). MoS2 is a well-known solid lubricant with a layered structure which can significantly reduce boundary friction. At lower temperatures, this was not observed because viscosity is higher, and therefore the oil plays a dominant role. It is also possible that there is thermal activation leading to the lubricity of MoS2, which is most likely related to the chemical interaction with the rubbing surfaces and formation of a tribochemical film. Formulations with both MoS2 and surfactant (Fig. 5f) demonstrated interference of the surfactant effect and the MoS2 effect. While at the lower temperatures of 20 and 40 °C the behavior is similar to the PAO 10 + BN + S#2, at 100 °C, the decrease in the coefficient of friction due to MoS2 appears to be counteracted by the surfactant, resulting in roughly the same coefficient of friction at all the temperatures. The surfactant stabilizes nanoparticle suspensions through the adsorption onto the nanoparticles, thus preventing agglomeration and settling of inorganic additives. The surfactant layer may also prevent or minimize interactions of MoS2 nanoparticles with rubbing surfaces, thereby hindering the formation of a tribofilm and resulting in higher coefficient of friction than MoS2 alone.

3.3 Wear

For pure PAO 10 and PAO 10 + BN + S#2, the skirt profile measurements were conducted after each of the 3 h friction tests at 20, 40, and 100 °C (Fig. 6a). The original skirt sample had sharp ridges that wore off during the friction tests. After a run-in period, during which the topmost asperities became worn, the same plateaus were observed in both lubricants at 20 and 40 °C. This result is most likely related to the high lubricant viscosity at these temperatures limiting the severe asperity interaction. At 100 °C though, clear differences in wear were seen between the samples tested with pure PAO 10 and PAO 10 + BN + S#2, while the same is true also for the rest of the nanoparticle formulations. The analyses of wear on the samples after 3 h of friction tests at 100 °C (9 h of total friction test time) are presented in Fig. 6b. The largest wear (12.5 μm) was observed with pure PAO 10 as a lubricant. Addition of only the surfactant resulted in a wear of approximately 3.0 μm. Addition of 3 wt% of BN nanoparticles resulted in 5.0 μm of wear with and without the surfactant. MoS2 appeared to provide the optimal protection against wearing off ridges, resulting in the wear off of approximately 2.0 μm with and without surfactant. The fact that the wear did not change in the nanoparticle formulations with and without surfactant indicates that the surfactant plays a minor role in the wear mechanism.
Fig. 6

Graphs showing 2D profilometric wear measurements of a skirt samples tested using PAO 10 and PAO 10 + BN + S#2 at 20, 40 and 100 °C, and b skirt samples tested in PAO 10 and PAO 10 containing 3 wt% BN or 3 wt% MoS2, with and without surfactant #2 at the end of 9-h-long tests

Optical microscope images of the worn liner surface taken under the same lighting conditions, after the friction tests (Fig. 7), show distinct differences in visual appearance. All the samples were rinsed with acetone and wiped dry before the microscope examination. The liner tested with pure PAO 10 is shown in Fig. 7a. It is apparent that the deeper machining marks are still present on the surface, while shallower machining marks began to disappear. This result is in agreement with the severe wear of the piston skirt segment in pure PAO 10 (Fig. 6a), confirming that PAO 10 offers little protection to both the rubbing parts. Figure 7b shows the surface tested, with only the surfactant being added to PAO 10. The blue hue of the surface is indicative of tribochemical film formation. The retention of the original machining marks is the evidence confirming the beneficial role of this tribofilm, which prevents excessive wear (Fig. 6). Chlorinated hydrocarbons are known for their extreme-pressure function; they can form scuffing-resistant tribofilms by thermal decomposition [15]. The appearance of the cast iron surface tested when using PAO 10 + BN as additive (Fig. 7c) is similar to that tested in the case of pure PAO 10 (Fig. 7a). As will be discussed later, a tribofilm does not seem to be forming in this case. It is interesting to note that, in the case of PAO 10 + BN + S#2 (Fig. 7d), the surface does not look similar to either that of Fig. 7b or that of Fig. 7c. It is possible that BN nanoparticles prevent the formation of a tribofilm, even though the surfactant has the tendency to form such film as was observed in Fig. 7b. That could be attributed to either a chemical mechanism or a physical mechanism during which BN particles may act as abrasive particles continuously removing it and preventing its coherent formation. Furthermore, binding of the surfactant molecules and the BN nanoparticles may prevent the interaction of surfactant with cast iron and the formation of a tribofilm. In tests with PAO 10 + MoS2, the formation of a tribofilm on the cast iron surface is obvious (Fig. 7e). The tribofilm, while it might not be entirely chemical in nature, is adhered well onto the surface, because rinsing with acetone could not remove it. The presence of the original machining marks denotes that little wear occurred, which is in agreement with the low coefficient of friction and with the low wear measured for the skirt sample (Fig. 6b). The appearance of the surface tested with PAO 10 + MoS2 + S#2 (Fig. 7f) is different from the case with just MoS2 (Fig. 7e). It resembles the surface tested with PAO 10 + BN + S#2 (Fig. 7d). It is possible that the surfactant prevents MoS2 nanoparticles from adhering to the surface of cast iron. Nonetheless, the presence of MoS2 still reduces the wear and friction at 100 °C.
Fig. 7

Microscope images of cast iron disks tested with a PAO 10, b PAO + S#2, c PAO 10 + BN, d PAO 10 + BN + S#2, e PAO 10 + MoS2, and f PAO 10 + MoS2 + S#2

3.4 Raman Spectroscopy

Tribochemical films and the near-surface material govern friction and wear behaviors. Tribochemical films are formed during lubricated sliding contact as a result of reaction between the surface material, lubricant additives, and the environment. The effect of nanomaterial and surfactant additives to PAO 10 was further investigated using Raman microscopy to obtain a better understanding of the reduced friction and wear mechanisms.

The Raman spectra of a clean cast iron surface, pure PAO 10, and nanolubricants with and without surfactant were obtained along with the spectra of the friction-tested cast iron samples (Fig. 8). The Raman spectrum of pure PAO 10 shows no specific Raman-sensitive features except for periodic intensity oscillation between 800 and 2,000 cm−1, which is typically attributed to vibration coupling of alkane and alkene groups in the long-chain PAO molecules. The PAO 10 spectrum changes when the nanoparticles are added. Specifically, four resonance bands appear in both BN and MoS2 nanolubricant formulations in the range of 800–1,600 cm−1. The position and intensity of those peaks are independent of the type of nanoparticle and the presence of surfactant (Figs. 8, 9), which therefore can be ascribed to some ordering of PAO molecules along the nanoparticle interface [16]. Peaks near 1,300 ± 10 cm−1 and 1,450 ± 10 cm−1 are attributed to in-phase wagging and deformation of (CH3) and (−CH2) groups; the series of peaks between 843 and 891 are attributed to the out-of-plane hydrogen modes for olefins (H–C=); and the bands between 1,064 and 1,079 cm−1 correspond to C=C=C stretching [17].
Fig. 8

Raman spectra of MoS2-based nanolubricants (3 wt%) with and without surfactant (marked as +S and (no S), respectively) and cast iron cylinder liners tested

Fig. 9

Raman spectra of nanolubricants (3 wt% BN or 3 wt% MoS2 in PAO 10) and cast iron surfaces tested

Hexagonal BN has only one Raman-active high energy band at 1,366 cm−1 [18], which is well defined in the spectrum of BN nanoparticles’ suspension in PAO 10 (Fig. 9a). However, this peak is absent on the spectrum of cast iron tested with BN nanolubricant (Fig. 9a). Minor peaks on this spectrum may be indicative of chemical interactions between BN and cast iron, but they cannot be attributed to any of the original compositions. Therefore, it can be concluded that BN is not present at the tested cast iron interface. This result correlates well with the frictional performance of the lubricant containing BN.

The Raman spectra of the MoS2-based nanolubricants with and without surfactant (Figs. 8, 9b) show bands in the range of 100–800 cm−1 that are characteristic of a MoS2 resonant Raman mode that occurs when the He/Ne laser excitation at 633 nm (1.96 eV) is used. Proximity of the excitation energy to the MoS2 semiconductor direct band gap (1.95 eV at room temperature [19]) causes strong coupling of energy into phonon modes resulting in rich multi-photon resonant Raman spectra, which can be used as a sensitive probe to monitor changes in the electronic states of the system caused by structural differences [20, 21]. The resonant Raman peaks and their origins described in the literature for a variety of MoS2 samples are summarized in Table 4.
Table 4

Summary of Raman and resonant Raman peaks of MoS2

Raman peak, cm−1

Description

Notes

Reference

32 (first order)

E2g2

Vibration of two rigid layers against each other

[19]

178

A1g(M)-LA(M)

 

[19, 20]

227

LA(M)

Longitudinal acoustic mode, appears only for nanomaterials not for single crystals from scattering of zone-boundary photons activated by facet disorder.

[19, 23]

286 (first order)

E1g

Vibration of atoms within S–Mo–S layer (not distinguishable in this study, most likely due to the random orientation of crystallites in powder sample).

[19, 20, 23]

383 (first order)

E2g1(G)

Vibration of atoms within S–Mo–S layer

[19, 20]

408 (first order)

A1g(G)

Vibration of atoms within S–Mo–S layer

[19, 20]

422

b sum of two phonon process non-Raman optical phonon

The Stokes scattering of the b band has been attributed to a two-phonon Raman process of successive emission of a dispersive quasi-acoustic (QA) phonon and a dispersionless transverse-optical (TO) phonon. Both propagate along the c axis.

[23]

453–464

A2u or 2 × LA(M)

The two peaks around 460 cm−1 were demonstrated to be a superposition of two peaks 453 and 464 cm−1, with enhancement of the 453 cm−1 feature in the spectra of nanoparticles [20].

[19, 23]

527

E1g(M) + LA(M)

Observed only for large >500 nm and bulk crystals

[19]

569

2 × E1g(G)

 

[19, 20]

598

E2g1(M) + LA(M)

 

[19]

642

A1g(M) + LA(M)

 

[19]

763

2 × E2g1

 

[23]

778

A1g(M) + E2g1(M)

 

[20]

822

2A1g(G) or 2A1g(M)

 

[20]

Four first-order modes (32, 286, 383, and 408 cm−1) appear at all excitation energies, while other peaks shown in Table 4 are present only on the resonant Raman spectra and correspond to the coupling of phonon modes to electronic transitions associated either with the band-gap or the exotic states (multi-peak states that violate the perturbative pulse splitting law). The relative intensity of peaks in different samples can be used for the interpretation of morphological differences. For example, it was demonstrated that the intensity of A1g peak (408 cm−1) compared to the intensity of E2g1 (383 cm−1) peak increases as the particle size decreases [22, 23]. Superimposing the two peaks at ~460 cm−1, the relative intensity of the feature at 453 cm−1 increases as the particle size decreases [22]. Also, additional low-frequency features observed at ~230 cm−1 in the nanocrystalline samples have been demonstrated to be characteristic of crystalline disorder, whereas they are absent in the spectra of fully crystalline samples [24].

The MoS2 resonance Raman peaks appear less intense in PAO 10 suspension compared with the tested cast iron sample, probably because of random orientation of the suspended nanoparticles and the scattering of PAO 10 molecules. It should be noted here that the MoS2 features are even less intense in the formulation with added surfactant. This could be due to better dispersion of nanoparticles in PAO 10 or stronger scattering of laser radiation, although no specific surfactant-related peaks were observed.

The difference between the spectra of the friction-tested cast iron samples with MoS2 nanolubricant with and without surfactant (Fig. 8) can be interpreted as a surfactant effect. Based on A1g and E2g1 peak ratios (Fig. 8), the size of MoS2 particles is larger in the film formed from the formulation with the surfactant. This conclusion based on smaller MoS2 particle sizes in the film formed without surfactant is confirmed by the increased contribution of 453 cm−1 peak in the spectrum without surfactant vs. higher contribution of 464 cm−1 peak when using MoS2 lubricant with the surfactant. It is possible that the surfactant prevents nanoparticles from degradation/shearing under severe condition of friction tests. Structural modifications of Mo–S–I nanowires under severe boundary lubrication conditions, which led to the formation of MoS2 sheets in the contact area, were previously correlated to lower coefficient of friction and reduced wear [25]. Therefore, the MoS2 tribofilm formed without surfactant compared to that formed with surfactant is in agreement with the lower coefficient of friction (Fig. 5e, f) and stronger interaction with the cast iron surface observed using optical microscopy (Fig. 7e, f). Additional information can be extracted from the appearance of 420 cm−1 line on both (MoS2 with and without the surfactant) cast iron spectra. According to previous studies [26], this line is indicative of the long range crystal orientation with the c axis parallel with the laser (lamellar sheets perpendicular to the laser), therefore maximizing the Raman cross section of this line. This line is only present on the surface of tested cast iron samples, indicative of the lamellar alignment of MoS2 particles, but not observed on the spectra of the nanolubricant because of the random orientation of MoS2 nanoparticles in suspensions (Fig. 8).

Windom et al. [27] argued that microcrystalline MoS2 powders tend to undergo oxidation to yield MoO3 (peaks at 279, 820, and 994 cm−1) when subjected to high laser power, friction tests, and temperatures. The bands around 820 cm−1 observed on our samples with the tribofilms are sharp and quite well resolved, similar to the spectra of natural MoS2 crystal and are not likely due to oxidation. The absence of the other two peaks characteristic of MoO3 (278 and 994 cm−1) also confirms this assumption. Therefore, no oxidation of MoS2 during the friction tests was observed.

4 Conclusions

Friction and wear studies carried out using formulations of PAO 10 oil with different additives at 20, 40, and 100 °C examined the effects of BN and MoS2 nanoparticles as well as the effect of surfactant on the tribological behavior of gray cast iron cylinder liner segments against an aluminum alloy piston skirt segments. The results showed that BN did not offer any improvement in friction under the tested conditions, while the addition of MoS2 nanoparticles was very effective in reducing both friction and wear compared to the base PAO 10 oil. The wear measurements of the skirt samples support the trends observed in the friction tests. Furthermore, the use of surfactant proved not only beneficial in suspending the nanoparticles in the solution, but also lowered the friction and wear by itself. The analysis of Raman spectra showed that MoS2 nanoparticles added to PAO 10 oil form an aligned MoS2 tribofilm, which assisted in significantly lowering the coefficient of friction and reducing the wear of the sample. Surfactant appears to partially suppress the interactions between MoS2 additive and the cast iron surface. BN nanoparticles, having lamellar structure similar to MoS2, added to PAO 10 oil did not lower the coefficient of friction, and were not detected on the surface of cast iron after the test.

Notes

Acknowledgments

This study is a part of Industrial Technology Program #M68008852 supported by the US Department of Energy. The scanning electron microscopy was performed at the Electron Microscopy Center for Materials Research, and Raman Microscopy was conducted at the Center for Nanoscale Materials, at Argonne National Laboratory—The US Department of Energy Office of Science Laboratory—operated under Contract No. DE-AC02-06CH11357 by the UChicago Argonne, LLC. The authors would like to thank Eduardo Tomanik at Mahle Metal Leve S.A. for providing the samples used in this study.

References

  1. 1.
    Nakayama, K., Yasutake, Y., Takiguchi, M., Furuhama, S.: Effect of piston motion on piston skirt friction of a gasoline engine. SAE paper 970839 (1997)Google Scholar
  2. 2.
    Demas, N.G., Erck, R.A., Fenske, G.R.: Tribological evaluation of piston skirt/cylinder liner contact interfaces under boundary lubrication conditions. Lubr. Sci. 22, 73–87 (2010)Google Scholar
  3. 3.
    Tung, S.C., McMillan, M.L.: Automotive tribology overview of current advances and challenges for the future. Tribol. Int. 37, 517–536 (2004)CrossRefGoogle Scholar
  4. 4.
    Etsion, I.: State of the art in laser texturing. J. Tribol. 127(1), 248–253 (2005)CrossRefGoogle Scholar
  5. 5.
    Wang, Y., Brogan, K., Tung, S.C.: Wear and scuffing characteristics of composite polymer and nickel/ceramic composite coated piston skirts against aluminum and cast iron cylinder bores. Wear 250, 706–717 (2001)CrossRefGoogle Scholar
  6. 6.
    Mosey, N., Muser, M., Woo, T.: Molecular mechanisms for the functionality of lubricant additives. Science 307, 1612–1615 (2005)CrossRefGoogle Scholar
  7. 7.
    Bhushan, B., Gupta, B.K.: Handbook of tribology. McGraw-Hill, New York (1991)Google Scholar
  8. 8.
    Sun, L., Tao, X., Zhao, Y., Zhang, Z.: Synthesis and tribology properties of stearate-coated Ag nanoparticles. Tribol. Trans. 53(2), 174–178 (2010)CrossRefGoogle Scholar
  9. 9.
    Erdemir, A.: A crystal chemical approach to the formulation of self-lubricating nanocomposite coatings. Surf. Coat. Technol. 200, 1792–1796 (2005)CrossRefGoogle Scholar
  10. 10.
    Campbell, S.E., Luengo, G., Srdanov, V.I., Wudi, F., Israelachvili, J.N.: Inorganic fullerene-like material as additives to lubricants: structure-function relationship. Nature 382, 520–522 (2006)CrossRefGoogle Scholar
  11. 11.
    Rapoport, L., Leshchinsky, V., Volovik, Y., Lvovsky, M., Nepomnyashchy, O., Feldman, Y., Popovitz-Biro, R., Tenne, R.: Modification of contact surfaces by fullerene-like solid lubricant nanoparticles. Surf. Coat. Technol. 163–164, 405–412 (2003)CrossRefGoogle Scholar
  12. 12.
    Dang, H., Sun, L., Zhou, J., Zhang, Z.: Synthesis, structure and tribological properties of stearic acid coated (NH4)3PMo12O40 nanoparticles. Tribol. Lett. 17, 311–316 (2004)CrossRefGoogle Scholar
  13. 13.
    Zhang, Z., Xue, Q., Zhang, J.: Synthesis, structure and lubricating properties of dialkyldithiophosphate-modified Mo–S compound nanoclusters. Wear 209, 8–12 (1997)CrossRefGoogle Scholar
  14. 14.
    Truhan, J.J., Qu, J., Blau, P.J.: The effect of lubricating oil condition on the friction and wear of piston ring and cylinder liner materials in a reciprocating bench test. Wear 259, 1048–1055 (2005)CrossRefGoogle Scholar
  15. 15.
    Kotvis, P.V., Tysoe, W.T.: Surface chemistry of chlorinated hydrocarbon lubricant additives—part I: extreme-pressure. Tribol. Trans. 41(1), 117–123 (1998)CrossRefGoogle Scholar
  16. 16.
    Larson, R.G.: The structure and rheology of complex fluids. Oxford University Press, New York (1999)Google Scholar
  17. 17.
    Miller, F.A., Mayo, D.W., Hannah, R.W.: Course notes on the interpretation of infrared and Raman spectra. Wiley-Interscience, New York (2004)Google Scholar
  18. 18.
    Nemanich, R.J., Solin, S.A., Martin, R.M.: Light-scattering study of boron-nitride micro-crystals. Phys. Rev. B 23, 6348–6356 (1981)CrossRefGoogle Scholar
  19. 19.
    Frey, G.L., Elani, S., Homyonfer, M., Feldman, Y., Tenne, R.: Optical-absorption spectra of inorganic fullerenelike MS2 (M=Mo, W). Phys. Rev. B 57, 6666–6671 (1998)CrossRefGoogle Scholar
  20. 20.
    Wieting, T.J., Verble, J.L.: Infrared and Raman studies of long-wavelength optical phonons in hexagonal MoS2. Phys. Rev. B 3, 4286–4292 (1971)CrossRefGoogle Scholar
  21. 21.
    Sekine, T., Uchinokura, K., Nakashizu, T., Matsuura, E., Yoshizaki, R.: Dispersive Raman mode of layered compound 2 h-MoS2 under the resonant condition. J. Phys. Soc. Jpn. 53, 811–818 (1984)CrossRefGoogle Scholar
  22. 22.
    Frey, G.L., Tenne, R., Matthews, M.J., Dresselhaus, M.S., Dresselhaus, G.: Raman and resonance Raman investigation of MoS2 nanoparticles. Phys. Rev. B 60, 2883–2892 (1999)CrossRefGoogle Scholar
  23. 23.
    Stacy, A.M., Hodul, D.T.: Raman spectra of IVB and VIB transition metal disulfides using laser energies near the adsorption edges. J. Phys. Chem. Solids 46, 405–409 (1985)CrossRefGoogle Scholar
  24. 24.
    McDevitt, N.T., Zabinski, J.S., Donley, M.S., Bultman, J.E.: Disorder-induced low-frequency Raman band observed in deposited MoS2 films. Appl. Spectrosc. 48, 733–736 (1994)CrossRefGoogle Scholar
  25. 25.
    Joly-Pottuz, L., Dassenoy, F., Martin, J.M., Vrbanic, D., Mrzel, A., Mihailovic, D., Vogel, W., Montagnac, G.: Tribological properties of Mo–S–I nanowires as additive in oil. Tribol. Lett. 18, 119–125 (2005)CrossRefGoogle Scholar
  26. 26.
    Wakabayashi, N., Smith, H.G., Nicklow, R.M.: Lattice dynamics of hexagonal MoS2 studied by neutron scattering. Phys. Rev. B 12, 659–663 (1975)CrossRefGoogle Scholar
  27. 27.
    Windom, B.C., Sawyer, W.G., Hahn, D.W.: A Raman spectroscopic study of MoS2 and MoO3: applications to tribological systems. Tribol. Lett. 42, 301–310 (2011)CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC (outside the USA) 2012

Authors and Affiliations

  • Nicholaos G. Demas
    • 1
  • Elena V. Timofeeva
    • 1
  • Jules L. Routbort
    • 1
  • George R. Fenske
    • 1
  1. 1.Argonne National LaboratoryArgonneUSA

Personalised recommendations