at SciVerse ScienceDirect

Energy 54 (2013) 167e173

Contents lists available

Energy

journal homepage: www.elsevier .com/locate/energy

Wear prevention characteristics of a palm oil-based TMP(trimethylolpropane) ester as an engine lubricant

N.W.M. Zulkifli a,*, M.A. Kalam a, H.H. Masjuki a, M. Shahabuddin a, R. Yunus b

aDepartment of Mechanical Engineering, University of Malaya, Kuala Lumpur 50603, Malaysiab Institute of Advanced Technology, University Putra Malaysia, 43400 Serdang, Selangor, Malaysia

a r t i c l e i n f o

Article history:Received 2 September 2012Received in revised form9 January 2013Accepted 14 January 2013Available online 17 February 2013

Keywords:Sliding wearBiolubricantCoefficient of frictionLubrication

* Corresponding author. Tel.: þ60 3 79675204; fax:E-mail address: nurinmz@um.edu.my (N.W.M. Zul

0360-5442/$ e see front matter � 2013 Elsevier Ltd.http://dx.doi.org/10.1016/j.energy.2013.01.038

a b s t r a c t

This paper presents the experimental results carried out to evaluate wear prevention characteristics ofa palm oil-based TMP (trimethylolpropane) ester using a four-ball machine for different regime oflubrication. The TMP ester is produced from palm oil, which is biodegradable and has high lubricityproperties such as a higher flash point temperature and VI (viscosity index). Three different regimes oflubrications are investigated, which hydrodynamic, elasto hydrodynamic and boundary lubrications.Under these test conditions, the wear and friction characteristics of different TMP samples are measuredand compared. For boundary lubrication, it is found that up to 3% addition of Palm oil-based TMP ester inOL (ordinary lubricant) decreases the maximum amount of WSD (wear scar diameter) and reduces (COFcoefficient of friction) up to 30%. Highest amount of load (220 kg) carrying capacity was also found fromthe contamination of 3% TMP. For hydrodynamic lubrication, addition of 7% of TMP reduces the frictionup to 50%. It is well known that mechanical efficiency of machinery component increases withdecreasing COF. The results of this investigation will be used to develop new and efficient lubricant tosubstitute the petroleum-based lubricant partially for automotive engine application.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

There has been enormous interest in the use of oils fromrenewable sources such as animal fats, vegetable oils and bio-lubricants [1e3]. The uncertainty of the crude oil supply, pollutantsemissions and its higher price give biodegradable biolubricantsmore advantages over mineral base oils. However, vegetable oil inits natural form has limited usage due to its poor oxidation stability[4] and inferior low temperature properties [5]. The first bio-degradable lubricants were developed for two-stroke outboardengines in the early 1980s, using neopentylpolyol esters ofbranched-chain fatty acids as based fluids [6]. Eychenne andMouloungui [7] have reviewed the developments in environ-mentally friendly lubricating oils based on neopentylpolyols suchas neopentyl glycol, pentaerythritol and trimethylolpropane. Otherresearches [8,9] have worked on engine oil prepared from a mix-ture of trimethylolpropane esters having both sufficiently highviscosity and low pour point. Jayadas et al. [10] found that coconutoil could be a good boundary lubricant as far as the coefficient of

þ60 3 79675317.kifli).

All rights reserved.

friction is concerned. But the wear rate is higher than ordinarylubricants, which may not be recommended as an engine oil in theunmodified form. Themajor function of engine oil is to decrease thefrictional force, by forming a thin film between the moving parts inthe engine such as piston and cylinder. Consequently, wear of themoving parts is reduced and diminished the power loss. Besides,engine oil also used to cool the engine by carrying away the heat,inhibits corrosion, cleans and improves sealing. One of the mostimportant properties of engine oil is viscosity, which should be highenough to maintain the lubricating film thickness, and low enoughto make sure the oil can flow through all the engine parts. The TMP(trimethylolpropane) ester is produced from a palm oil methyl esterthrough transesterification. Transesterification eliminates thehydrogen molecule on the beta carbon position of the palm oilsubstrate, thus improving the oxidative and thermal stability of theTMP ester; an important lubricity property in vegetable oils [11]. Inaddition, TMP esters have good friction-reducing properties andacceptable anti-wear properties [12]. However, in contrast, Waaraet al. [13] found that pure synthetic ester (TMP-Oleate and TMP-C8-C10) in a sliding contact resulted in high wear rates and a lot ofabrasive marks on the surface. To understand the lubricity perfor-mance, the Stribeck diagram can be seen in Fig. 1. This diagramshows the coefficient of friction in a rotating bearing, plotted

Nomenclature

OL ordinary lubricantTMP trimethylolpropanePOME palm oil methyl estersVI viscosity indexCOF coefficient of frictionWSD wear scar diameterHFRR high frequency reciprocating rig

Fig. 2. Synthesis of palm oil-based TMP ester.

N.W.M. Zulkifli et al. / Energy 54 (2013) 167e173168

against Stribeck variable (known as the dimensionless Sommerfieldnumber). The Stribeck variable is shown in equation below.

Stribeck variable; ¼mNP

; m ¼ viscosity;

N ¼ rotational speed; P ¼ unit load

Lubrication under low values of Stribeck region is known asboundary lubrication. Low values of Stribeck variable correspond tolow viscosity, low speeds, and high loads. In this case, the surfacesasperities are actually in contact to each other [14]. Therefore,extreme pressure and anti wear additives played an important roleto form a boundary lubricating film [14e16] to protect the surface.In addition, surface characteristic also plays an important rolecontacting two surfaces. There are many locations in the moderndiesel engine where boundary lubrication occurs such as in pistonrings at the top and bottom dead centers, and the fuel-injectionsystem. Lubricity is the fluid property that characterizes frictionand wear phenomenon in boundary lubrication. The lubrication inthe large values of the Stribeck is called hydrodynamic region. Largevalues of the Stribeck variable correspond to high viscosity fluids,high rotational speeds, and light loads. Under these conditions, anoil film supports the load/carry the load at low friction. In thiscondition, fluid properties are of greatest significance for wear andfriction, and the surface characteristics such as surface asperities ofthe bearing are not important since the surfaces are not in contact[17,18]. In an engine, hydrodynamic lubrication occurs in journalbearing, crankshaft bearing and when the piston rings are moving[19,20]. It can be highlighted that frictional forces consume about15e20% of fuel in an internal combustion engine. Only thedevelopment of low coefficient of friction lubricant and surfacedevelopment is the solution to reduce this energy consumption.Hence, in order to have a better understanding of reduction offrictional forces through the tribological study, a new type oflubricant substitute TMP is introduced. The effect of TMP ester on

Fig. 1. Stribeck curve diagram.

different lubrication regime using four-ball machine and HFRR(high frequency reciprocating machine) wear tests have beeninvestigated. The output of these experimental results would givea correlation of actual frictional behavior in internal combustionengine. The objectives of this study are to investigate the tribo-logical properties under boundary and hydrodynamic lubricationwhen added palm-oil based TMP ester in OL (ordinary lubricant).

2. Experimental method

2.1. Lubricant sample preparation

Palm oil-based TMP ester was synthesized by the trans-esterification of methyl esters as shown in Fig. 2 [21]. The prepa-ration process of palm oil-based TMP ester is shown in Fig. 3. Thetrimethylolpropane [2-ethyl-2-(hyhydroxymethyl)-1,3-propane-diol] was chosen due to its lower melting point compared to otherpolyols. A 200 g volume of palm oil methyl ester and a knownamount of TMP were placed into a 500 ml three-neck reactor andconstantly agitated by a magnetic stirrer. The weight of TMP wasdetermined based on the required molar ratio and the calculatedmean molecular weight of the POME (palm oil methyl esters). Themixture was then heated to the reaction temperature in the pres-ence of catalyst. A vacuumwas gradually applied to the system untilthe desired pressure was reached. This pressure was maintaineduntil completion of the reaction. The peaks appeared were identi-fied and labeled based on the number of alkyl carbon groups thatattached to TMP backbone. The esters formed are identified bymaking comparisons by standard or by using the standard of TG(triglycerides), DG (diglycerides) and monogliceride (MG) [22].Fatty acid composition is shown in Table 1.

Both palm oil-based TMP esters were blended with OL usinga stirrer at 110 rpm and heated to 100 �C. Detail properties of OL areshown in Table 2. The blended lubricants consisted of 5%, 10%, 15%,20% and 100% palm oil TMP esters (volume basis) with OL. Com-positions of the lubricant samples are shown in Table 3.

Palm Oil Methyl Ester Trimethylolpropane

Heating and stirring of mixture Palm Oil Methyl Ester and Trimethylolpropane

Alkaline catalyst was added to mixture

Product purification: using vacuum distillation

A vaccum gradually applied

Purified palm oil-based TMP ester

Fig. 3. Preparation of palm oil-based TMP ester.

Table 1Fatty acid composition of palm oil-based TMP ester.

Fatty acid composition (%)

Monoester 5Diester 5Triester 90%

Table 3Percentages of the palm oil-based TMP ester and OL in each sample.

Sample Palm oil-based TMP ester (%) OL (%)

OL 0 100TMP5 5 95TMP10 10 90TMP15 15 85TMP20 20 80TMP100 100 0

N.W.M. Zulkifli et al. / Energy 54 (2013) 167e173 169

2.2. Lubricant properties test

Density: Density was measured using a DMA 35 portable den-sity meter. The samples were tested at 15 �C. Only 2 ml of eachsample was used per test.Viscosity index (VI): The viscosity index (VI) was calculatedusing ASTM D2270. This method uses the kinematic viscositiesat 40 �C and 100 �C to determine the VI. The experiments werecarried out using an ISL automatic Houillon viscometer.

2.3. Wear and friction test

2.3.1. Four-ball machineThe four-ball wear test is used to investigate the effect of Palm

oil-based TMP ester under wear preventive test (hydrodynamictest). The four-ball wear tester is the predominant wear tester usedby the oil industry to study lubricant chemistry. The four-ball weartester consists of three balls held stationary in a ball pot plusa fourth ball held in a rotating spindle as shown in Fig. 4. The ballsused in this study were steel balls, AISI 52-100, 12.7 mm in diam-eter, with 64e66 Rc hardness. These balls were thoroughly cleanedwith toluene before each experiment. The sample volume requiredfor each test was 10 ml. The test method used to investigate wearpreventive characteristic was ASTM D4172 [23]. The test conditionswere 40 kg load, operating temperature of 75�C � 2 �C, rotationalspeed of 1200 rpm and operation time of 60 min. For the extremepressure test, ASTM 2783 [24] where the load is increased by 20 kgfor every 10 s until the ball is welded. The wear produced on thethree stationary balls is measured under a calibrated microscopeand reported as the WSD (wear scar diameter) or calculatedvolume.

2.3.2. HFRR (High frequency reciprocating rig) configurationThe HFRR wear test was used to investigate the effect of Palm

oil-based TMP ester under fluid film lubrication as shown in Fig. 5.The HFRR test (ASTMD6079) used in this study involves aweightedcast iron pin cylinder (6 mm length) and a stationary cast iron plat(15 mm � 15 mm), which is completely submerged in a 10 ml ofsample. The ball and disk is tested at room temperature andbrought into contact with each other and the entire apparatus isvibrated at 10 Hz for 60minwith the load 10 N. The sliding stroke ismaintained at 2 mm. Several blends of Palm oil-based TMP esterincluding 1%, 3%, 5%, 7% and 10% are tested and analyzed. Each testwas carried out three times repeatedly to observe any errors needto be analyzed.

Table 2Properties of ordinary lubricant.

Properties Ordinary lubricant

Density at 15 �C (g/cm3) 0.867Pour point (�C) �36Flash point (�C) 228TBN (mg KOH/g) 8.6

3. Results and discussion

3.1. Physicochemical properties of palm oil-based TMP ester in OL(ordinary lubricant)

The data were used to evaluate the differences between OL andto serve as a basis for comparing the blended fuels with palm oil-based TMP esters. The percentage of palm oil-based TMP estersand OL in each sample is shown in Table 2. All of the lubricantproperties are listed in Table 4.

The VI (viscosity index) of oil is a number that indicates theeffect of temperature changes on its viscosity. A low VI signifiesa relatively large change in viscosity with changes in temperature.In other words, the oil becomes extremely thin at high tempera-tures and extremely thick at low temperatures. A high VI signifiesrelatively little change in viscosity over a wide range of tempera-tures. The ideal oil for most purposes is one that maintains a con-stant viscosity throughout different temperature changes.Considering automotive lubricants can easily show the importanceof the VI. The oil having a high VI resists excessive thickening whenthe engine is cold and, consequently, promotes rapid starting andprompt circulation. It also resists excessive thinning when the en-gines sliding components are hot and thus provides full lubricationand prevents excessive oil consumption. The VI of oil is determinedbased on known viscosities at any two temperatures (here 40 �Cand 100 �C). It can be seen from Table 3, TMP 1 has the highestviscosity as 108.75 cSt.

The density of a lubricant fluid can provide indications of itscomposition and nature. The density of lubricants, mainly hydro-carbons, varies between 0.860 and 0.980 g m�3. The densities of OLand TMP100 were 0.850 and 0.9012, respectively.

3.2. Effect of palm oil-based TMP ester under wear preventive test(hydrodynamic test)

The WSD results for the different percentages of palm oil-basedTMP ester in OL are shown in Fig. 6. The significance improvementin WSD was found for TMP3, which is 30% as compared to OL. Thiscan be explained that an increase in the number of ester groupsleads to greater binding of the molecules, which provide a greater

Fig. 4. A schematic of the four-ball test machine.

Fig. 5. The schematic diagram of HFRR wear test.

N.W.M. Zulkifli et al. / Energy 54 (2013) 167e173170

resistance to shear forces. However increasing TMP ester more than3% does not show any improvement onWSD. This can be attributedto the fact that, more than 3% of Palm oil-based TMP ester could notform the stronger lubricant film formation.

This finding is similar to the findings reported by Yunus et al.[25], who found the lower wear scar diameter using palm oil-basedTMP ester with compared to ordinary hydraulic fluid (not the en-gine lubricant). In addition, Masjuki et al. [26] also found that palm-based lubricating oil had a better wear performance compared tomineral oil. Fernandez et al. [27] reported that the addition ofa synthetic TMP ester to a low viscosity polyalphaolefin acted asa wear reducer. According to Havet et al. [28] the length of the fattyacid chains tends to increase the adsorbed film thickness, thereforeincreasing the surface area protected.

The values of COF (coefficient of friction) for the various lubri-cant samples are shown in Fig. 6. It has similar trend like WSD. Ingeneral it can be stated that the COF reduces with the addition ofPalm oil-based TMP ester except for TMP100. Based on the value ofCOF and WSD it is found that TMP 3 has the best lubricity perfor-mance. At TMP 100,WSD ismaximumof around 0.78mm. Based onFig. 7, it can be seen that after the operation time of 20min, the thinfilm layer has started breaking down. This is because of low vis-cosity in TMP 100, the lubricant film formed is not able to supportthe load throughout the operation time.

The micrographs of the area around the wear scar and the wornsurfaces of the ball specimens for the different percentages of thePalm oil-based TMP esters are shown in Fig. 8. It can be seen that,adhesive wear occurs. The load applied to the contacting asperitiesthat they deform and adhere to each other to form micro-joints.

Table 4Properties of the different percentages of the palm oil-based TMP ester in OL.

Sample Properties

Viscosity VI Density (g/cm3)

40 �C (cSt) 100 �C (cSt)

OL 101.86 � 1.53 14.46 � 0.35 146 0.8549 � 0.0174TMP 1 108.75 � 1.22 15.08 � 0.28 145 0.8553 � 0.0168TMP 3 107.21 � 1.41 15.11 � 0.21 148 0.8558 � 0.0176TMP 5 101.42 � 1.30 14.89 � 0.15 153 0.8562 � 0.0191TMP 7 102.15 � 1.17 14.78 � 0.10 150 0.8568 � 0.0176TMP 10 97.637 � 1.21 14.43 � 0.11 153 0.8583 � 0.0165TMP100 40.03 � 1.32 9.15 � 0.11 221 0.9012 � 0.0171

The continuous rotation of the top ball causes the rupture of themicro-joints and creates pits and grooves. Surface morphologyshows that adhesive is very minimal for OL as we can see there istransferred material.

For lubricant added Palm oil-based TMP ester, the appearance ofthe worn surfaces seems to present corrosive product of black colorat higher temperature. The higher percentage of palm oil-basedTMP esters added, the black color is darker. This may beexplained from the fact that at higher temperature, lubricant can beoxidized and thereby produces different types of corrosive acidsthat enhance corrosivewear [29,30]. In addition, it is found that thisboundary lubricating films acts as a sacrificial layer. The reactionproducts of tribochemistry provide a layer, so instead of the surfacebeing worn, this layer is removed [31]. It can be seen that, fromFig. 8(TMP 10), the black color areas have a severe wear comparedto non-black area. Further it can be explained that the existence ofoxygen might accelerate the generation of inorganic oxides likeFe3O4, Fe2O3, which plays an important role in the formation oflubrication film on the rubbing surface [32]. For TMP 100, it seemslike the fluid film of lubricant is completely breakdown. The directinteraction between two solids that remove material from, while atthe same time producing polished surface on [33]. Based fromFig. 8(TMP100), material removal in this case occurs as the verysmall slabs of surface steel ball being lifted out the surface, probably

Fig. 6. Wear scar diameters (WSD) and coefficient of friction (COF) for different per-centages of the palm oil-based TMP esters in OL. Maximum error WSD; �0.05 mm,maximum error COF: �0.005.

Fig. 7. Relationship between friction torque and times for different percentages of thepalm oil-based TMP esters in OL. Maximum error of friction torque: �0.01 Nm.

Fig. 8. SEM micrograph from the stationary ball of different percentages of the palmoil-based TMP ester in OL.

N.W.M. Zulkifli et al. / Energy 54 (2013) 167e173 171

by some form of delamination. The polished wear mechanism oc-curs here as a result of severe abrasivewear. It can be seen also fromFig. 8 (TMP100) that, the light reflected from real surfaces.

This phenomenon was confirmed with the results of surfaceroughness measurement, as shown in Fig. 9. The surface roughnesstest was conducted in order to understand the wear scar texture onthe bottom three balls. A stylus profilometry test was performed toevaluate the surface texture of the worn halo on the rotating balland the wear scars on the stationary balls. The profilometry tracewas fed into a microprocessor and the resulting signal was dis-played on a chart recorder providing surface roughness graphs. Theprofilometer traces were taken perpendicular to the direction ofsliding. Sample TMP 7 showed a comparatively smooth wear pat-tern compared to the other samples. No significant difference isobserved among the tested lubricant samples except TMP 100,which shows the maximum surface roughness of 0.07.

3.3. EP (Extreme pressure) characterization

Under extreme pressure, the wear rate from the rubbing surfacecan be minimized by reducing the bearing stress or enhancing thestrength. At extreme pressure the flash temperature is elevated andthe extreme pressure additives decompose to form a chemical re-action film which tends to increase the bearing area and con-sequently reduce the contact pressure. The performances of the EPadditives are dependent on three factors, which are: strength of thetribo-film, reaction rate between the additives and the rubbingsurface, and finally the compatibility between the base oil and theadditives [34]. To evaluate the friction and wear characteristics ofdifferent biolubricants the extreme pressure characteristics wereinvestigated for a load range of 20 kg up until the balls specimensweld with each other, which is defined as the FSL (final seizureload). The rotational speed applied was 1200 rpm, the temperaturewas 27 � 7 �C and an operation period of 10 s was employed foreach sample.

Fig. 10 shows the variation in the coefficient of friction underdifferent loads for different TMP ester added lubricant. The mag-nitude of the coefficient of friction indicates that the lubricationregime occurring in the rubbing zone were both elasto hydrody-namic and boundary lubrications. Most of the TMP ester presenteda lower coefficient of friction (COF) compared to the OL. It seemsquite clear that the COF for TMP 1, TMP 3, TMP 5 and TMP 7 werelower as compared to petroleum-based ordinary lubricant whileTMP 10 and TMP 100 showed the higher COF throughout the loadrange. The OL, TMP 1 and TMP 3 were found to be more stable interms of a FSL (final seizure load) which is defined as the load atwhich lubricants film totally break down and testing ball materialsbecome welded. It can also be noted that TMP 1 imparted lowest

Fig. 9. Surface roughness for the different lubricating oil samples. Error ¼ �0.01 mm.Fig. 11. Relationship of friction with time under hydrodynamic lubrication.Error ¼ �0.005 Nm (Reciprocating test).

N.W.M. Zulkifli et al. / Energy 54 (2013) 167e173172

COF upto initial seizure load (ISL) of the load at which a consid-erable amount of wear is occurred however, after FSL the lowestCOF was found to be using TMP 3. The results of Fig. 10 imply thatthe TMP 1 and TMP 3 have the greatest ability to retain its prop-erties upto the load of 220 kg without the lubricating film breakingdown. This can be attributed to the fact that at higher loads thelubricating film thickness becomes thinner than some of the as-perities present in the boundary lubrication regime. However,these asperities are covered by the long chain fatty acid and theesters of the biolubricants, which are known as surface activematerials. TMP 1 and TMP 3 showed the best performance in termsof reducing the COF, which can be interpreted as the fatty acidpresent in the TMP esters acting as an active surface material whichis adequate for these two lubricants, and above that the fatty aciddoes not take part because the rubbing contact surface is fullycovered by the them. Another important fact is that above 3%addition the TMP adversely affects the quality of the lubricant.

3.4. Effect of palm oil-based TMP ester under fluid film lubrication

According to results presented in Fig. 11, TMP 7 has showed thelowest friction torque around 0.02 Nm. The OL shows the highestfriction torque because the viscosity is too high and created a thickfilm, which will cause more friction. Addition of Palm oil-basedTMP ester in the OL reduced the viscosity of the lubricant, henceprovides a better thin film at the surface and thus reduced thefriction. Based on the experimental results, it can be stated that, theaddition of Palm oil-based TMP ester in the base lubricant leads to

Fig. 10. Variation in coefficient of friction at different load under extreme pressurecondition.

improve the frictional torque characteristics of the lubricants. Incontrast with boundary lubrication, at low load, TMP 100 still canmaintain the film lubrication. The weight loss of the cast iron flatplate is very small. It can be assumed almost no wear occur in thisregime lubrication.

4. Conclusion

The experimental results show that Palm oil-based TMP esterbased lubricant improves the wear preventive lubrication prop-erties in terms of COF and WSD. The TMP 3 shows the lowest WSDand COF with compared to others lubricants analyzed in this study.In addition, under extreme pressure condition maximum loadbearing capacity of 220 kgwas found to be using TMP 3 by retainingits quality without breakdown. Under the fluid film lubricationusing reciprocating test the TMP 7 shows the best lubricity propertywith the lowest friction torque. In order to utilize Palm oil-basedTMP ester as an engine oil, many other properties including oxi-dative, thermal and hydrolytic stabilities need to be examined.Palm oil-based TMP ester is environmentally desired tomineral oil-based lubricants, research to investigate the properties of palm oil-based TMP ester to make it technologically competitive as auto-mobile lubricant, and should be encouraged.

Acknowledgments

The authors would like to thank the University of Malaya, whichmade this study possible through the research grant FRGS FP020/2011A, PV070/2011B and high impact research grant no UM.C/HIR/MOHE/ENG/07.

References

[1] Rhee I-S. Evaluation of environmentally acceptable hydraulic fluids. NLGISpokesman 1996;60:28e35.

[2] Sharma BK, Adhvaryu A, Erhan SZ. Friction and wear behavior of thioetherhydroxy vegetable oil. Tribol Int 2009;42:353e8.

[3] Waara P, Hannu J, Norrby T, Byheden Å. Additive influence on wear andfriction performance of environmentally adapted lubricants. Tribol Int 2001;34:547e56.

[4] Erhan SZ, Sharma BK, Perez JM. Oxidation and low temperature stability ofvegetable oil-based lubricants. Ind Crops Prod 2006;24:292e9.

[5] Adhvaryu A, Erhan SZ, Perez JM. Wax appearance temperatures of vegetableoils determined by differential scanning calorimetry: effect of triacylglycerolstructure and its modification. Thermochim Acta 2002;395:191e200.

[6] Van der Waal G, Kentebeek D. Environment protective ester based fluids. In:Proceedings of Tribology 2000, 8th International Colloquium. Germany:Technische Akademie Esslingen; 1992. p. 1e8.

N.W.M. Zulkifli et al. / Energy 54 (2013) 167e173 173

[7] Eychenne V, Mouloungui Z. Neopentylpolyol ester-based lubricants. Ole-agineaux Corps Gras Lipides;3:57e63.

[8] Leleu G, Bédague P, Silion B. Trimethylolpropane esters useful as base lubri-cants for motor oils. Patent US. Institut Francais du Petrole, Rhone Progil;1977.

[9] Mattei L, Pacor P, Piccone A. Oils with low environmental impact for moderncombustion engines. J Synth Lubr 1995;12:171e89.

[10] Jayadas NH, Prabhakaran Nair K, A G. Tribological evaluation of coconut oil asan environment-friendly lubricant. Tribol Int 2007;40:350e4.

[11] Yunus R, Fakhru’l-Razi A, Ooi TL, Omar R, Idris A. Preparation and character-ization of trimethylolpropane esters from palm kernel oil methyl esters. JPalm Oil Res;15:42e9.

[12] Shah S, Moser BR, Sharma BK. Glycerol tri-ester derivatives as diluent toimprove low temperature properties of vegetable oils. J ASTM Int 2012;7.

[13] Waara P, Norrby T, Prakash B. Tribochemical wear of rail steels lubricated withsynthetic ester-based model lubricants. Tribol Lett 2004;17:561e8.

[14] Hsu SM, Gates RS. Boundary lubricating films: formation and lubricationmechanism. Tribol Int 2005;38:305e12.

[15] Hamrock BJ. Fundamentals of fluid film lubrication. Nueva York, EUA:McGraw-Hill; 1994.

[16] Hsu SM. Boundary lubrication: current understanding. Tribol Lett 1997;3:1e11.[17] Hunter WB, Zienkiewicz OC. Effect of temperature variations across the

lubricant films in the theory of hydrodynamic lubrication. J Mech Eng Sci1960;2:52e8.

[18] Pinkus O, Sternlicht B. Theory of hydrodynamic lubrication. New York:McGraw-Hill; 1961.

[19] Hronza J, Bell D. A lubrication and oil transport model for piston rings usinga NaviereStokes equation with surface tension. SAE Technical Paper 2007-01-1053; 2007.

[20] Sorab J, Korcek S, Bovington C. Friction reduction in lubricated componentsthrough engine oil formulation. SAE Trans 1998;107:1903e10.

[21] Yunus R, Fakhrulrazi A, Ooi TL, Iyuke SE, Idris A. Preparation and character-ization of trimethylolpropane esters from palm kernel oil methyl esters. J OilPalm Res 2003;15:42e9.

[22] Yunus R, Lye O, Fakhru’l-Razi A, Basri S. A simple capillary column GC methodfor analysis of palm oil-based polyol esters. J Am Oil Chem Soc 2002;79:1075e80.

[23] ASTM D4172 -94 Standard test method for wear preventive characteristics oflubricating fluid (four-ball method). West Conshohocken, PA: ASTM Interna-tional; 2010.

[24] ASTM D2783 -03 Standard test method for measurement of extreme-pressureproperties of lubricating fluids (four-ball method). West Conshohocken, PA:ASTM International; 2009.

[25] Yunus R, Fakhru’l-Razi A, Ooi TL, Iyuke SE, Perez JM. Lubrication properties oftrimethylolpropane esters based on palm oil and palm kernel oils. Eur J LipidSci Technol 2004;106:52e60.

[26] Masjuki HH, Maleque MA, Kubo A, Nonaka T. Palm oil and mineral oil basedlubricants-their tribological and emission performance. Tribol Int 1999;32:305e14.

[27] Fernandez Rico JE, Hernandez Battez A, Garcia Cuervo D. Wear preventioncharacteristics of binary oil mixtures. Wear 2002;253:827e31.

[28] Havet L, Blouet J, Robbe Valloire F, Brasseur E, Slomka D. Tribological charac-teristics of some environmentally friendly lubricants. Wear 2001;248:140e6.

[29] Briscoe BJ, Thomas PS, Williams DR. Microscopic origins of the interfacefriction of organic films: the potential of vibrational spectroscopy. Wear 1992;153:263e75.

[30] Klaus EE, Nagarajan R, Duda JL, Shah KM. The adsorption of tribochemicalreaction products at solid surfaces. In: Proceedings of the Institution of Me-chanical Engineers; 1987. p. 379e87.

[31] Quinn TFJ, Sullivan JL, Rowson DM. Origins and development of oxidationalwear at low ambient temperatures. Wear 1984;94:175e91.

[32] Lu X, Cotter J, Eadie DT. Laboratory study of the tribological properties offriction modifier thin films for friction control at the wheel/rail interface.Wear 2005;259:1262e9.

[33] Samuels LE, Consultants S. Polishing wear. In: ASTM handbook: friction,lubrication, and wear Technology, vol. 18; 1992. p. 191e8.

[34] Lin Y-C, Cho Y-H, Chiu C-T. Tribological performance of EP additives in dif-ferent base oils. Tribol Trans 2011;55:175e84.