Analysis of thermal stability and lubrication characteristics of Millettia pinnata oil

Abstract

Lubricants are mostly used to reduce the friction and wear between sliding and metal contact surfaces, allowing them to move smoothly over each other. Nowadays, due to the increase in oil prices and reduction of oil reserves, it is necessary to replace mineral oil, which will also protect the environment from hazards caused by these oils. It is essential to find an alternative oil for the replacement of mineral-oil-based lubricants, and vegetable oil already meets the necessary requirements. Vegetable-oil-based biolubricants are non-toxic, biodegradable, renewable and have a good lubricating performance compared to mineral-oil-based lubricants. This study analyzes the thermal stability and lubricating characteristics of different types of vegetable oil. The friction and wear characteristics of the oils were investigated using a four-ball tester, according to ASTM method 4172. Millettia pinnata oil has good oxidation stability due to the presence of higher percentages of oleic acid in its fatty acid composition. Millettia pinnata oil also shows a higher kinematic viscosity. Rice bran oil shows a higher viscosity index than other oils, and it is better for boundary lubrication. In thermogravimetric analysis, it was found that Millettia pinnata oil remains thermally stable at 391 °C. Millettia pinnata oil showed a lower coefficient of friction and rice bran oil showed a lower wear scar diameter compared to other vegetable oils and lube oils. A lower wear scar surface area was found with rice bran oil compared to other vegetable and commercial oils. Therefore, due to a better lubricating performance, Millettia pinnata oil has great potential to be used as a lubricating oil in industrial and automotive applications.

---

1. Introduction

In this world, a large number of lubricants are used as lube oils, i.e., refined oils, synthetic oils, mineral oils, and vegetable oils. Mostly, mineral oils are used as lubricant oils in engine oil and motor oil. Mineral oils are available in the world market, and these are derived from petroleum oil, but these oils have negative effects on the environment due to their non-biodegradability and toxicity.^1,2 Biolubricant oil should be used as an alternative to mineral oil due to the depletion and increased prices of crude oil, and in the interests of environmental protection. Vegetable oils are renewable, less toxic, biodegradable, and thermally stable compared to mineral and commercial oils.^1,3,4 Biolubricants have a better lubrication performance due to their higher viscosity, viscosity index, flash and fire points.^5,6 High frictional coefficients, higher pour points, and poor oxidation stability are the main disadvantages of vegetable oils.⁷ There are long fatty-acid chains and polar groups in vegetable oils; therefore, these oils can be used as biolubricants for boundary and hydrodynamic lubrication purposes.^8,9 Vegetable oils are obtained from various kinds of oil seeds, and almost 350 oil-bearing crops are available all over the world. Vegetable oils are found in both edible and non-edible sources, such as jatropha, Millettia pinnata, olive, palm, coconut, sunflower, Moringa oleifera, Calophyllum inophyllum, canola, and soybean.¹⁰ Many researchers used vegetable oils as an alternative fuel for diesel engines, but some researchers reported that biolubricants, which are produced from crude vegetable oils or derived from vegetable oils, can be used as lubricating oils.

Lubrication is required for reducing the frictional effects and wear associated with various types of sliding, moving, and rotating components.^11,12 It can also reduce the temperature in metal contact zones. Friction is one of the main causes of loss of energy. It has been stated that one-third of the total world energy is lost due to friction between sliding and moving parts. The quality and type of the lubricating oil plays an important role in reducing friction between moving and sliding components. An extensive amount of energy is used to reduce friction, especially in the power-generation, industrial, and transportation sectors, and the main economic losses are caused by wear in mechanical parts and transmission systems, and their mechanisms and replacement.¹³ In transportation sectors, 33% of energy losses are caused by friction in piston assemblies, crankshaft and camshaft mechanisms, valve trains, pumping, bearings and transmission systems.¹⁴ Lubricants are mostly used in industrial and engine oil applications.¹⁵ Many researchers reported that vegetable oils, i.e., coconut, olive, sunflower seed, corn, soybean, rice bran, and rubber seed oil, can be used as biolubricants for industrial and engine-lubrication purposes. Rani et al.¹⁶ evaluated the lubrication and tribological characteristics of rice bran oil, which was used as a biolubricant oil for industrial applications. They reported that rice bran oil has good thermal properties and tribological characteristics. The viscosity and oxidation stability of rice bran oil can be developed by the addition of proper additives. Suitable additives, such as organozinc compounds, aromatic amine compounds, and sulphur phosphorus compounds, were mostly used for improving the oxidation stability of the oils. Aravind et al.¹⁷ investigated the lubricating properties of rubber seed oil, and found that this oil can be used as a lubricant under cold conditions due to the higher viscosity index and low pour point of rubber seed oil.

Millettia pinnata (karanja or Pongamia pinnata) oil is one of the family members of Leguminosae. It is found as a native in the Northern part of Australia, India, some regions of Eastern Asia, and Fiji. It has a high potential for oil production, at 200 thousand metric tons per year all over the world.^18,19 India is the largest producer of Millettia pinnata oil, with a yearly production of 135 000 metric tons per year. The maximum height of a medium-sized tree is 18 m at an altitude of 1200 m. The Millettia pinnata tree is inherent to humid and subtropical surroundings, and it grows well in areas having an annual rainfall ranging from 500 to 2500 mm. In its natural habitat, the temperature ranges from a maximum of 27–38 °C to a minimum of 1–16 °C.^20,21 It grows up quickly, matures after 4–7 years and starts yielding fruits, which are elliptical, flat and long (7.5 cm). Millettia pinnata kernels contain fatty oil (27.5%), moisture (19.0%), starch (6.6%), protein (17.4%), ash (2.4%) and crude fiber (7.3%). Every single tree has a yield of 9–90 kg seeds per year, and there is potential for the production of 900–9000 kg seed per ha. The seed contains almost 30–40% oil. A thick yellow-orange to brown, bitter, non-drying, non-edible oil is extracted from the seeds. Generally, 18–22% oil is extracted by the village crushers and 24–27.5% oil is extracted by Indian mills.²² The extracted oil is mostly used for the production of soap, illuminating oil, and tanning leather. The presence of high contents of triglycerides in the oil makes its odor and taste unpleasant, because of bitter falconoid elements, such as pongamia and karanja.²³ This oil is also used as a water-paint binder, pesticide, and lubricant. Millettia pinnata oil contains a high oleic acid content (44.5–71.3%) as the main fatty acid, along with palmitic acid (3.7–7.9%), stearic acid (2.4–8.9%), and linoleic acid (10.8–18.3%). On the other hand, Millettia pinnata oil also contains almost 9.5–12.4% eicosenoic acid (9-eicosenoic acid). This dark-brown oil has a repulsive odor and shows fungicidal properties.²⁰

The aim of this study is to evaluate the thermal stability and lubrication characteristics of Millettia pinnata oil as a biodegradable lubricant for industrial purposes. This work also includes investigation of the lubrication properties, performance analysis of vegetable oils, and their comparison with lube oil. This study presents the biolubricant as an alternative lubricating oil in industrial and automotive applications. This work also illustrates vegetable oil properties, DSC, TGA, wear and friction analysis.

2. Materials and methods

2.1. Materials

Millettia pinnata and rice bran oil were collected from Indian markets by personal communication with foreign suppliers. The oil was extracted from Millettia pinnata seeds, which contained 30–40% oil. The refined Millettia pinnata oil is very clear, odorless, and acidic. Coconut oil and SAE40 were collected from local suppliers in Malaysia.

2.2. Determination of fatty acid composition

The fatty acid composition is the most important characteristic of vegetable oils when these are used as lubricating oils. The coefficient of friction (COF) and wear is reduced due to the presence of long carbon chains in the fatty acid composition. The FACs of vegetable oils were measured using gas chromatography (GC) analysis, according to the ASTM D5555 method. In GC analysis, a small amount of oil (0.02 g) was diluted with hexane (1.5 ml) in a small vial. The diluted solution was charged with a flame ionization detector, which was connected to a GC analyzer. Each peak was identified and then compared with the FAC of a standard mixture.

2.3. Determination of oil properties

Every lubricating oil has some important properties, such as viscosity, viscosity index, density, cloud point, pour point, acid value, flash point, fire point, and oxidation stability. The density of the oil was measured by using a Stabinger viscometer (SVM 3000) according to the ASTM D7042 method. The lubricating properties, such as saponification value, iodine value, and acid number, are affected by the chemical characteristics of vegetable oils. All chemical properties were measured for different types of vegetable oil, according to ASTM standard methods. The amount of alkali compounds, which are required to saponify all the triglycerides of the oil, is measured by the saponification value. The saponification value is defined as the amount of potassium hydroxide (KOH) in mg required to saponify 1 gram of the oil. The iodine value represents the number of unsaturated bonds contained in the oil. The iodine value is the amount of iodine absorbed per gram of the oil. The acid numbers of different types of vegetable oils were measured using an automated titration system with an accuracy of ±0.001 mg KOH g⁻¹, by following the ASTM D664 method. The oxidation stability of vegetable oils was measured according to the EN 14112 method by using 873 Rancimat equipment. The dynamic and kinematic viscosity were measured with a Stabinger viscometer (SVM 3000) according to ASTM D7042, at 40 °C and 100 °C. The kinematic viscosity results were obtained automatically, and are equivalent to the ASTM D445 method. The viscosity index was estimated by using the ASTM D2270 method. The flash point and fire point temperature were measured with a Pensky–Martens flash point tester and Cleveland open cup equipment, according to ASTM D92 and 93 methods, respectively. The pour point and cloud point temperature were measured with cloud and pour point testers, according to ASTM D97 and D2500 methods, respectively. The oil crystallization temperature was investigated by digital scanning calorimetry (DSC). A TA DSC (Q200) was used for DSC analysis with temperatures ranging from −90 °C to 550 °C. The thermal mass degradation of different types of vegetable oils and SAE40 were conducted with a thermogravimetric analyzer (TGA). In TGA analysis, a TGA Q500 instrument was used with a temperature range from 0 °C to 600 °C and TGA analysis of the oil was performed under nitrogen gas at a flow rate of 20 L min⁻¹, while the temperature increase was maintained at 10 °C min⁻¹.

2.4. Friction and wear test procedure

The wear and friction tests were conducted with a TR-30H four-ball tribotester. Chrome alloy steel balls were used as testing balls in this experiment. The characteristics of wear and friction for different types of vegetable oil and SAE40 were tested according to the ASTM D4172 method. First of all, the tested steel balls were washed with n-heptane and then wiped with tissue to ensure that they were dry to use. These dry balls were placed in a steel cup and 10 ml tested oil was poured into the cup. In the experimental setup, three balls were kept stationary and another ball was rotating on these stationary balls. The fourth ball was adjusted within a collet and placed in a rotating arm. Winducom 2008 software was used for recording the frictional torque data. Fig. 1 shows a schematic diagram of a four-ball tribotester and Table 1 shows the experimental test conditions and testing ball configuration. After completing the test, the stationary balls were collected to evaluate the WSD of the steel balls and for performing SEM/EDX analysis. After obtaining the frictional torque data from the computer software, the COF of the tested oil was determined from this data.

Fig. 1 Schematic of four-ball tester.

Table 1 Experimental test conditions of four-ball tribotester

	Parameters	Condition
Operating conditions	Applying load	40 kg
	Speed	1200 rpm
	Fuel temperature	75 °C
	Duration	3600 seconds
	Materials	Chrome alloy steel (SKF)
	Metal composition	10.3% C, 0.08% S, 1.43% Cr, 0.11% P, 0.42% Mn, 0.05% Ni, 0.47% Si, 2.12% Zn, and remainder 85.02% Fe
Testing ball	Size	φ 12.7 mm
	Hardness	62 HRc
	Roughness of the surface	0.1 μm (C.L.A)
	Tensile strength	325 000 psi
	Yield strength	295 000 psi
	Density	7.85 gm cm^−3

---

2.5. Coefficient of friction (COF) calculation

The product of the spring constant and friction torque is called the coefficient of friction (COF). The frictional torque was measured with a load cell. In this experiment, the lower ball produced the maximum torque. The COF can be expressed as:where frictional torque can be expressed as T in kg mm, the applied load can be expressed as W in kg, and the distance can be expressed as r in mm. The distance (d) was measured from the center of the lower ball contact surface to the axis of rotation, and it was found to be 3.67 mm.

2.6. Wear evaluation

According to the ASTM D4172 method, the wear scar diameters (WSD) of the tested balls were evaluated using an optical microscope (C2000, IKA, UK) with a ±0.01 mm resolution. The magnifying lens was adjusted to a better position on the ball scar surfaces to obtain a very clear scar image. Computer software was used for capturing the scar images and also the WSD was measured with this software. After completing the measurement of the WSDs of the tested balls, the average WSD was calculated from the obtained results. After completing the WSD measurements, these tested balls were collected for SEM/EDX analysis.

2.7. Statistical and error analysis

For statistical analysis, a two-tailed paired t-test was used for independent variables, which showed a significant difference among the sample set means, using Microsoft Excel 2013. Differences among the mean values at a level of p = 0.05 (95% confidence level) were considered statistically significant. Error analysis was required to determine the percentage of uncertainty and accuracy of the experimental results. The experimental test results had errors and uncertainties due to the selection of instruments, environmental conditions, test conditions, test planning, observation, calibration, and reading. The experimental test results were collected at least three times and average values were calculated, using graph plotting and by measuring very precise data. Table 2 shows the accuracy of instruments and the relative uncertainty of various test parameters for the experiments. Sample calculations are provided in Appendix A. According to Monirul et al.,¹² the overall uncertainty was calculated at ±3.11% for these experiments. The overall uncertainty was calculated as follows:

Table 2 Summary of relative uncertainty and accuracy

Parameter	COF	WSD
Accuracy	±0.5	±0.01
Relative uncertainty (%)	±2.24	±2.16

---

3. Results and discussion

3.1. Fatty acid composition

The fatty acid compositions (FACs) of the vegetable oils are shown in Table 3. Millettia pinnata oil has a higher degree of unsaturation compared to other vegetable oils, except for coconut oil, which contains an almost 91% saturated FAC.²⁴ It was observed that Millettia pinnata contains a 28.2% saturated, 49.7% monounsaturated, and 21.7% polyunsaturated FAC. On the other hand, rice bran oil has a 23.6% saturated, 42.6% monounsaturated, and 31.5% polyunsaturated FAC. Millettia pinnata oil contained 48.3% oleic acid compared to other vegetable oils, and can be used as a lubricant base stock. The presence of a higher oleic acid content in Millettia pinnata oil can enhance the lubrication performance by reducing the friction at contact surfaces.²⁵ While rice bran oil contained 42.6% oleic acid, it can be used as a lubricant. Rani et al.¹⁶ reported that the higher percentage of oleic acid (41.86%) contained in rice bran oil allows it to be developed as a lubricating oil. It was reported that the presence of a higher content of oleic acid in vegetable oil allows it to be used as a lubricant base stock, as a promising substitute for mineral and synthetic oil-based lubricating oils.²⁵

Table 3 Fatty acid compositions of different types of vegetable oils

Carbon structure	Fatty acid name	Coconut	Rice bran	Millettia pinnata
C8:0	Caprylic	9	—	—
C10:0	Capric	7	—	—
C12:0	Lauric	49	—	—
C14:0	Myristic	16	0.3
C16:0	Palmitic	8	20.7	12.7
C16:1	Palmitoleic	—	—	0.1
C18:0	Stearic	2	2.6	11.9
C18:1	Oleic	6	42.6	48.3
C18:2	Linoleic	2	31.5	19.5
C18:3	Linolenic	—	—	2.2
C20:0	Arachidic	—	—	1.6
C20:1	Eicosenoic	—	—	1.3
C22:0	Behenic	—	—	1.4
C24:0	Lignoceric	—	—	0.6
Saturated		91	23.6	28.2
Monounsaturated		6	42.6	49.7
Polyunsaturated		2	31.5	21.7

---

3.2. Properties of vegetable oils

Table 4 shows the physical and chemical properties of different types of vegetable oil. A higher density was found in rice bran oil (924.1 kg m⁻³) than coconut oil (918 kg m⁻³) and Millettia pinnata oil (918.6 kg m⁻³). The iodine value, saponification value and acid number are the main chemical properties of lubricant-based oils; these properties can be enhanced to give a better lubricating performance. The unsaturated FAC can be evaluated from the iodine value, which indicates the percentage of unsaturated fatty acids contained by the oil. Millettia pinnata oil showed a lower iodine value (87.5 g l₂/100 g) compared to rice bran oil (97 gl₂/100 g) besides coconut oil having the lowest value (8 gl₂/100 g); oils with lower iodine values are often good lubricants. Vegetable oil should have a lower acid number for a better lubricating performance. Coconut oil has a lower acid number (2.8 mg KOH g⁻¹) than other vegetable oils. The acid number of the oil can determine the base content, which is required to neutralize the lubricating oil. The presence of free fatty acids in the oil can be characterized by the acid number. The free fatty acid content of the oil increases with the acid number. A higher acid value was found for Millettia pinnata oil (39.67 mg KOH g⁻¹) than coconut (2.8 mg KOH g⁻¹) and rice bran oil (4.31 mg KOH g⁻¹). This result can be attributed to the higher level of free fatty acid present in the oil, and it has some limiting factors with regard to its use as a biolubricant oil. It is clear that Millettia pinnata oil contains a higher amount of free fatty acid compared with coconut and rice bran oil. Nevertheless, some methods are used to de-acidify the Millettia pinnata oil, such as physical refining, chemical esterification, or conventional alkalinification.²⁶ The saponification value can measure the amount of alkali compounds that are required for conversion of the oil into soap.²⁷ A lower saponification value indicates that an oil has a lower tendency to saponify at high temperatures. Rice bran oil has a lower saponification value (182 mg KOH g⁻¹) than coconut oil (258 mg KOH g⁻¹) and Millettia pinnata oil (184.3 mg KOH g⁻¹). The saponification value is one of the objectionable properties of vegetable oils when they are considered as lubricating oils at high temperatures.¹⁶

Table 4 Physical and chemical properties of vegetable and lube oils

Properties	Coconut	Rice bran	Millettia pinnata
Density (gm m^−3) at 40 °C	918	924.1	918.6
Iodine value (g l2/100 g)	8	97	87.5
Saponification value (mg KOH g^−1)	258	182	184.3
Acid value (mg KOH g^−1)	2.8	4.31	39.67
Oxidation stability (110 °C, h)	13.2	4.40	4.85

---

The lubrication performance depends on the oxidation stability of the oil, which may change its physical and chemical properties. A poor oxidation stability is the main disadvantage of vegetable-oil-based biolubricants. The presence of bis-allylic carbons between the two bonds is one of the main reasons for poor oxidation stability of the oil. The oxidation stability of the oil can be improved by adding some additives and by performing chemical modifications, such as epoxidation and hydrogenation.^28,29 It was reported that a higher percentage of unsaturated fatty acids (such as linoleic acid and linolenic acid) can reduce the oxidative stability of the oil, but the presence of highly saturated fatty acids provides a better oxidative stability, which may cause the oil to have a higher pour point.³⁰ It was found that the oxidation stabilities of coconut, rice bran, and Millettia pinnata were 13.20 h, 4.40 h, and 4.85 h, respectively. Millettia pinnata oil showed a better oxidation stability than other oils due to the presence of a higher percentage of oleic acid, gamma oryzanol, and natural antioxidant compounds. Millettia pinnata oil contained a higher amount of oleic acid, which may cause high oxidation stability compared to other oils. Anwar et al.³¹ reported that the presence of more oleic acid in oil is ideal to increase the oxidation stability. Therefore, Millettia pinnata oil has the potential to be used as a biolubricant. Viscosity is the most important property for lubricating oil to obtain a better lubrication performance. The rheological properties of vegetable oils and SAE40 are shown in Table 5.

Table 5 Rheological properties of vegetable and lube oils

Properties	Coconut	Rice bran	Millettia pinnata	SAE40
Kinematic viscosity at 40 °C (cSt)	26.327	39.225	44.175	103
Kinematic viscosity at 100 °C (cSt)	5.863	8.393	8.385	16.09
Viscosity index	164	193	169	148

---

The thermal properties of vegetable oils and lube oil are shown in Table 6. The pour point is the most important property for the lubricating oil, when it is used in the winter and in cold countries. A lower pour point temperature is better for good lubrication. Rice bran oil has a lower pour point (−13 °C) and cloud point (−6 °C) compared to other vegetable oils, while coconut oil has the highest cloud point (25 °C) and pour point (21.5 °C). The cloud and pour point temperatures were found to be 2 °C and −4 °C, respectively, for Millettia pinnata. The presence of a higher polyunsaturated FAC is the main reason for the lower pour and cloud point of rice bran oil. The pour point and cloud point properties of the oil could be influenced by the saturated, monounsaturated, and polyunsaturated fatty acid compositions.^32,33 The flash point is the temperature of the oil that indicates its volatility and safe operating temperature. The fire point indicates the temperature at which the oil is ignited by an external flame. The flash point and fire point temperatures can determine the volatility and fire resistance. Coconut oil possesses a higher flash point (321.5 °C) and fire point (325 °C) compared to rice bran oil (302.6 °C and 307 °C) and Millettia pinnata oil (227 °C and 232 °C). SAE40 has a lower flash point (204.8 °C) and fire point (207 °C) compared to vegetable oils. It was reported that the physical and chemical properties of SAE40 are completely dissimilar to vegetable oils because it is formulated as a lubricant from mineral oil. Mineral oil is a combination of C20–C50 hydrocarbons, which are usually known as naphthenic and paraffinic hydrocarbons. On the other hand, vegetable oils contain long-chain fatty acids (such as COOH groups), which are associated with a glycerol molecule, whereas no fatty acids are present in SAE40.

Table 6 Thermal properties of vegetable oils

Properties	Units	Coconut	Rice bran	Millettia pinnata	SAE40
Cloud point	°C	25	−6	2	−18
Pour point	°C	21.5	−13	−4	−23
Flash point	°C	321.5	302.6	227	204.8
Fire point	°C	325	307	232	207

---

3.2.1. Viscosity. Viscosity is the most important property with respect to identifying the individual grades of the lubricating oil. A higher viscosity indicates that the lubricant is being deteriorated by either contamination or oxidation, and a lower viscosity value indicates a decrease in the dilution properties of the oil.³⁴ The main disadvantage of vegetable-oil-based biolubricants is that they have a limited range of viscosity.³⁵ The kinematic viscosities of different types of vegetable oil and SAE40 at 40 °C and 100 °C are shown in Fig. 2. Millettia pinnata oil showed a higher kinematic viscosity (44.175 cSt at 40 °C) than other vegetable oils, except SAE40. These vegetable oils have satisfied International Standard Organization (ISO) requirements for their use as biolubricants. A lower viscosity produces more wear and a higher viscosity can cause more friction loss in moving and sliding metal components.³⁶ The viscosity index implies changes in viscosity with temperature. As can be seen from Table 5, rice bran oil possesses a viscosity index of 193, higher than coconut oil (164), Millettia pinnata (169), and SAE40 (148). A higher viscosity index indicates a smaller variation in viscosity with temperature, while a lower viscosity index indicates high viscosity changes with temperature.³⁷ Vegetable-oil-based biolubricants showed higher viscosity indices compared to commercial and mineral oils and, for this reason, they can be formulated in multi-grade lubricant applications. All vegetable oils exhibited viscosity indices above 145, making them suitable for a wide range of temperatures. These results can be attributed to the presence of triglyceride compounds in the vegetable oils, as these sustain stronger intermolecular interactions when the temperature is rising. Hence, Millettia pinnata oil can reduce wear at metal contact surfaces more effectively than other oils, and this is good for boundary lubrication applications.

Fig. 2 Viscosities of various vegetable oils at different temperatures.

3.3. Differential scanning calorimetry (DSC) analysis

Differential scanning calorimetry (DSC) analysis has been effectively used to evaluate the crystallization behavior of oils by analyzing the exothermic changes associated with this process.^38,39 The long-chain fatty acid compounds of the oils in the solid state exist in more than one crystalline form and thus have multiple melting points. The melting points of triacylglycerol depend on the carbon chain length, the nature of unsaturation, and the position and number of the carbon double bonds.⁴⁰ A DSC curve provides qualitative and quantitative information about the physical and chemical changes of the oil, and this curve shows endothermic (heat absorption) and exothermic (releasing heat) processes, as well as changes in the heat capacity of the oil.⁴¹ DSC analyses for vegetable oils are shown in Fig. 3. The DSC peaks indicate variations in the physicochemical properties of the oil during the heating process. The endothermic onset temperatures were found to be 8.13 °C, −20.06 °C, and −26.16 °C for coconut, rice bran, and Millettia pinnata, oil, respectively. Millettia pinnata oil showed a lower endothermic peak temperature due to the nature of the fatty acid constituents, namely oleic and linoleic acid. Coconut, rice bran, and Millettia pinnata oil showed melting temperatures of 22.29 °C, −12.08 °C, and −13.38 °C respectively, in the endothermic peaks. The enthalpy changes were found to be 89.08 J g⁻¹, 43.06 J g⁻¹, and 43.55 J g⁻¹ for coconut, rice bran, and Millettia pinnata oil, respectively. These results can be attributed to the presence of unsaturated triacylglycerol groups in the oil.⁴² It was seen that SAE40 oil showed a linear line; no melting temperature was found for this oil. It was reported that Millettia pinnata oil showed a good performance at low-temperatures, due to a lower saturated fatty acid composition and hydroxyl groups in the fatty acid chain, which may obstruct the crystal packing of TAG molecules.^33,43

Fig. 3 DSC analyses for vegetable oils.

3.4. Thermogravimetric analysis (TGA)

The thermal stability is the most important parameter of the oil, when it is used as a lubricating oil at high temperatures and when it is formulated from vegetable oils.^44,45 Thermogravimetric analysis has been used to observe the thermal stability of the oil. Thermogravimetric analyses (TGA) of vegetable oils are shown in Fig. 4–7 and Table 7. It was seen that all vegetable oils had better thermal stability than mineral-oil-based lubricants. Thermal stability has been determined by the onset temperature of decomposition, and it can be defined as the starting temperature of the decomposition of the oil. Thermal stability mainly depends on the chemical structure and FAC of the oil.⁴⁶ The decomposition temperatures were found to be 379.63 °C, 384.78 °C, 391.17 °C, and 249.20 °C with coconut, rice bran, Millettia pinnata, and SAE40 oil, respectively. Millettia pinnata oil showed a higher decomposition temperature than the other oils. The presence of a higher unsaturated FAC is the main reason for the high thermal stability of the oil.⁴⁴ The mass degradation of Millettia pinnata oil was compared with other vegetable oils. Almost 99% weight loss was found for all the vegetable oils, but 96% weight loss was found with SAE40. It is clear from the graph that 96% weight loss was detected for SAE40 over the temperature range 249–445 °C. The other oils, such as coconut, rice bran, and Millettia pinnata oil, were thermally stable at 379 °C (35 min), 384 °C (36 min), and 391 °C (37 min), respectively, and also 99% weight loss was found for each oil over the temperature ranges 379–509 °C, 384–502 °C, and 391–496 °C, respectively. The weight loss for Millettia pinnata oil was 9.844 mg between 286 °C and 496 °C, 9.743 mg for coconut oil between 262 °C and 509 °C, 14.60 mg for rice bran oil between 306 °C and 502 °C and 6.038 mg for SAE40 between 105 °C and 397 °C. The thermal stability of vegetable oil is higher due to the presence of a higher percentage of oleic acid and the addition of some additives to the oil.⁴⁷ Rani et al.¹⁶ reported that rice bran oil was more stable up to 37 min with no mass degradation. The decomposition temperature of an oil mainly depends on its free fatty acid composition and chemical structure. The thermal stability of an oil is mainly influenced by the chain length of fatty acids, branching, and degree of unsaturation.⁴⁸ Hence, Millettia pinnata oil has good thermal stability compared to other oils; the presence of a higher amount of oleic acid is the main reason for these results.

Fig. 4 TGA analysis for coconut oil.

Fig. 5 TGA analysis for rice bran oil.

Fig. 6 TGA analysis for Millettia pinnata oil.

Fig. 7 TGA analysis for SAE40 oil.

Table 7 Thermal stability data for some vegetable oils

Parameters	Coconut	Rice bran	Millettia pinnata	SAE40
Weight loss (mg)	9.743	14.6	9.844	6.038
% Weight loss	98.97	99.22	98.15	96.93
Temperature range at weight loss	262–509	306–502	286–496	105–397
Decomposition temperature (°C)	379.63	384.78	391.17	249.2
Residue (% per mg)	0.4141	0.3083	0.3156	0.4828
Weight of residue (mg)	0.04077	0.04537	0.05425	0.03007

---

3.5. Friction characteristics

Lubricant oil can create and maintain a stable lubricating film at the metal contact zone; it is the main ability of all lubricating oils. Vegetable oils provide an excellent lubricating performance due to their ester functionality.^49,50 Fig. 8. shows the variation of the coefficient of friction (COF) of vegetable oils with time (s), under a 40 kg load at 75 °C. A significant decrease in the COF at the start of each experiment was commonly observed in all oils. The COF of vegetable oils reached a steady-state condition after 30 seconds for each oil. It was clear that Millettia pinnata oil showed a lower COF than other vegetable oils. Because Millettia pinnata oil contains a higher unsaturated and a lower saturated FAC, this may cause a lower COF compared to other oils. Vegetable oils showed the lowest COF compared to commercial lubricant oil. SAE40 produced more friction when it ran for a longer period, showing a higher COF after 53 min. On the other hand, a lower COF was produced by Millettia pinnata after a long time period. The average frictional coefficients for different type of oils are shown in Fig. 9. The average COFs were 0.0529, 0.0621, 0.0296, and 0.0806 for coconut, rice bran, Millettia pinnata, and SAE40 oil, respectively. The average COF of Millettia pinnata oil was 44.04%, 52.33%, and 63.27% lower than those of coconut, rice bran, and SAE40, respectively. Millettia pinnata oil showed a lower average COF compared to other vegetable oils and commercial lube oil due to its higher viscosity. It is reported that vegetable oil contains triglycerides, which are a mainly polar group of ester molecules. The chain length of triglycerides can enhance the strength of a lubricating film and reduce the frictional effect from contacting surfaces. The presence of triglyceride components is the primary reason for the lower oxidation stability of vegetable oils.²⁴ Vegetable oils contain a higher FAC, which allows them to reduce the friction at a metal contact surface. It was reported that the carbon chain length of the FAC may reduce and control the frictional characteristics of the oil. Coconut oil contained a lower FAC with a lower carbon chain length (C8:0–C14 : 0), and it produced more friction at sliding contact surfaces compared to other vegetable oils. Millettia pinnata oil showed lower frictional characteristics compared to other vegetable oils due to the presence of long carbon chains in the FAC. Millettia pinnata oil can reduce more friction from sliding contact surfaces due to the presence of free fatty acids.^51,52 The friction increased as the acid value of the oil decreased, while wear was decreased. Vegetable oils have a higher acid value due to the presence of higher percentages of free fatty acids.⁵³ The frictional coefficient has been decreased due to the higher unsaturated fatty acid composition in the oil. Millettia pinnata oil contains a higher amount of oleic acid, which can reduce the friction. However, Millettia pinnata oil showed low frictional characteristics compared to SAE40; these results can be attributed to the free fatty acid content. Fatty acids contain carboxylic groups (COOH), which have the ability to adsorb on metallic surfaces to form a layer with the polar head adhering to the metal surface.^24,54

Fig. 8 Variation of COF for vegetable oils with respect to time.

Fig. 9 Average COF of vegetable oils.

3.6. Wear scar diameter

A higher temperature of the lubricant may cause more wear at sliding and metal contact surfaces.⁵⁵ The variation in the wear scar diameter (WSD) for vegetable oils is shown in Fig. 10. The average WSDs of Millettia pinnata oil were 13.96%, 27.90%, and 10.02% higher than coconut, rice bran, and SAE40, respectively. Millettia pinnata oil showed a higher WSD (0.792 mm) compared to other vegetable oils. A smaller WSD (0.571 mm) was found for rice bran oil and this has the maximum ability to protect metal to metal contact surfaces. The WSD of rice bran oil is lower compared to other vegetable oils and commercial lube oil. These results can be justified by the presence of natural anti-oxidants (gamma oryzanol and tocopherols) in the vegetable oils. The fatty acid chain length has a tendency to increase the absorbed film thickness, and if the length of the fatty acid is increased, the protective area of the metal contact surfaces also increases.⁵⁶ Wear has decreased with an increase in the unsaturated fatty acid composition in the oil. Oxidants are produced due to the fatty acid composition of the oil.⁵¹ However, rice bran oil has the maximum capability to maintain the lubricating film and interrupt the particles of wear by decreasing the specific wear between the contact areas to avoid the metal surface interactions. Millettia pinnata and coconut oil showed a higher wear rate due to their higher saponification values. Due to the moisture content in the air at high temperatures, coconut oil saponifies and produces undesirable products, which may cause more wear between the sliding parts.^57,58 Rice bran oil showed the lowest WSD compared to other oils, demonstrating the potential of rice bran oil as a possible lubricant ingredient.

Fig. 10 WSD of vegetable oils.

3.7. Wear scar surface analysis

In a mechanical system, there are several types of wear produced at various sliding and metal contact surfaces, such as adhesive, abrasive, corrosive, and fatigue wear. Wear increases due to the formation of peroxides at higher temperatures. In this experiment, the lubricant regime occurred by the boundary lubrication process, thereby, abrasive, adhesive, corrosive, and fatigue wear were observed at the metal rubbing zone.^59,60 Mostly, adhesive and abrasive wear were found in this experiment. The reduction of the lubricant film thickness brings the surfaces closer to each other, which may cause higher wear. Fig. 11 shows SEM images of wear scar surfaces for different types of oil. In this figure, three conditions of wear scar surfaces are shown for each oil. Additional surface deformation was found for rice bran and SAE40 oil, compared to other vegetable oils. The surface morphology showed worn surface metal layers, which were caused by the rotating balls moving in a sliding direction. It was reported that when the surface area was greater than 20 μm, it was damaged because of adhesive wear.⁶¹ It was observed that a lot of wear debris was present at the coconut-oil-lubricated surfaces. Some cracks and microcracks were present at surfaces lubricated by rice bran oil. It was found that coconut and rice bran oil can reduce more wear compared to other vegetable oils. It was seen that the wear track of the commercial lubricant oil was very smooth compared to the vegetable-oil-based biolubricant. More wear was found for vegetable oils than lubricating oil; mostly, these lubricated surfaces encountered oxidative or chemical wear. It is reported that the wear of the lubricated surfaces for each oil is caused by corrosion delamination. A smaller wear scar was found for rice bran oil compared to other vegetable oils and SAE40. These results can be attributed to the absorption of biodiesel on the metal surfaces because of the ester functionalities, which exist along with long carbon–hydrogen chains in fatty acid residues. Black spots were found at the lubricated surface of rice bran oil (a), Millettia pinnata oil (a) and SAE40 (a), which reflect the occurrence of oxidation caused by corrosion. The oxidation occurred due to the chemical reaction between different types of fatty acids and peroxides, which can have a negative effect on the lubricity.⁶² In some studies, it was reported that the wear characteristics of Millettia pinnata oil can be improved by adding different types of anti-wear additives, such as tricresyl phosphate (TCP), zinc dithiophosphate (ZDP), and zinc dialkyldithiophosphate (ZDDP).⁶³ The synergistic effect of organic molybdate ester with zinc dialkyldithiophosphate as an anti-wear additive has been established for rapeseed oil.⁶⁴ Another study showed that tyrosine derivatives could be used as an effective anti-wear additive for rapeseed oil.⁶⁵

Fig. 11 SEM images for different types of vegetable oil.

4. Conclusion

The thermal stability, lubricant properties, and characteristics of vegetable oils, such as coconut, rice bran, and Millettia pinnata oil, were compared to those of commercial lubricating oil. In general, all oils have unique lubricating properties that make them suitable for lubricant applications.

(1) Millettia pinnata oil contained almost 48.3% oleic acid and its oxidation stability is better than other vegetable oils and lube oil. Therefore, it has the ability to integrate multi-grade lubricating oil.

(2) Millettia pinnata oil showed a higher kinematic viscosity compared to other vegetable oils, and it is more suitable for boundary lubrication. Rice bran oil showed a higher viscosity index compared to other vegetable oils. For this reason, it can enhance the lubricating performance in lubrication applications. Vegetable-oil-based biolubricants have a higher viscosity index, and are very effective at high temperatures as they can maintain the lubricating film.

(3) Millettia pinnata oil is thermally more stable compared to other vegetable oils and lube oil. A lower pour point and cloud point were found with rice bran oil due to its higher polyunsaturated fatty acid composition.

(4) The average co-efficient of friction of Millettia pinnata oil is less compared to other vegetable oils and commercial oil because it contains a higher unsaturated fatty acid composition, which can reduce friction at metal contact surfaces. Rice bran oil showed a lower WSD than other oils.

(5) Rice bran oil showed a lower wear scar surface than other vegetable oils and the wear properties of Millettia pinnata oil can be improved by adding anti-wear additives. Therefore, Millettia pinnata oil has the potential to be used as a bio-based lubricant for industrial and commercial purposes.

Appendix: A

Tables 8 and 9

Table 8 Statistical analysis for WSD

Oil	Test ball 1	Test ball 2	Test ball 3	Mean	Variance	Standard deviation
Coconut	0.73066	0.71386	0.60048	0.681667	0.005014	0.07081
Rice bran	0.55429	0.55009	0.60888	0.571087	0.001076	0.032797
Millettia pinnata	0.84403	0.81884	0.71384	0.792237	0.004768	0.069052
SAE40	0.73906	0.70126	0.69835	0.71289	0.000516	0.022711

---

Table 9 Uncertainty analysis for WSD

Oil	Test 1	Test 2	Test 3	Max	Min	Accuracy		Average	Uncertainty (%)
						0.01	−0.01		+	−
Coconut	0.73066	0.71386	0.60048	0.73066	0.60048	0.74066	0.59048	0.66557	3.779587	−3.779587
Rice bran	0.55429	0.55009	0.60888	0.60888	0.55009	0.61888	0.54009	0.579485	2.072608	−2.072608
Millettia pinnata	0.84403	0.81884	0.71384	0.84403	0.71384	0.85403	0.70384	0.778935	3.356923	−3.356923
SAE40	0.73906	0.70126	0.69835	0.73906	0.69835	0.74906	0.68835	0.718705	1.832177	−1.832177
Uncertainty level (%)									2.760324	−2.760324

---

Acknowledgements

The authors would like to thank the University of Malaya for financial assistance through research grant numbers: CG 060-2013, FP 051-2014B and RP 016-2012B.

References

1. A. Adhvaryu, Z. Liu and S. Erhan, Ind. Crops Prod., 2005, 21, 113–119.
2. L. S. Khuong, N. W. M. Zulkifli, H. H. Masjuki, E. N. Mohamad, A. Arslan, M. H. Mosarof and A. Azham, RSC Adv., 2016, 6, 66847–66869.
3. S. Stolte, S. Steudte, O. Areitioaurtena, F. Pagano, J. Thöming, P. Stepnowski and A. Igartua, Chemosphere, 2012, 89, 1135–1141.
4. M. H. Mosarof, M. A. Kalam, H. H. Masjuki, A. M. Ashraful, M. M. Rashed, H. K. Imdadul and I. M. Monirul, Energy Convers. Manage., 2015, 105, 617–629.
5. S. Z. Erhan and S. Asadauskas, Ind. Crops Prod., 2000, 11, 277–282.
6. S. M. Alves, B. S. Barros, M. F. Trajano, K. S. B. Ribeiro and E. Moura, Tribol. Int., 2013, 65, 28–36.
7. L. Quinchia, M. Delgado, T. Reddyhoff, C. Gallegos and H. Spikes, Tribol. Int., 2014, 69, 110–117.
8. A. E. Atabani, A. S. Silitonga, H. C. Ong, T. M. I. Mahlia, H. H. Masjuki, I. A. Badruddin and H. Fayaz, Renewable Sustainable Energy Rev., 2013, 18, 211–245.
9. A. Adhvaryu, S. Z. Erhan and J. M. Perez, Wear, 2004, 257, 359–367.
10. H. Mobarak, E. N. Mohamad, H. Masjuki, M. Kalam, K. Al Mahmud, M. Habibullah and A. Ashraful, Renewable Sustainable Energy Rev., 2014, 33, 34–43.
11. S. C. Tung and M. L. McMillan, Tribol. Int., 2004, 37, 517–536.
12. I. M. Monirul, H. H. Masjuki, M. A. Kalam, M. H. Mosarof, N. W. M. Zulkifli, Y. H. Teoh and H. G. How, Fuel, 2016, 181, 985–995.
13. K. Holmberg and A. Erdemir, FME Transactions, 2015, 43, 181–185.
14. K. Holmberg, P. Andersson, N.-O. Nylund, K. Mäkelä and A. Erdemir, Tribol. Int., 2014, 78, 94–114.
15. A. Ahmed, H. Masjuki, M. Varman, M. Kalam, M. Habibullah and K. Al Mahmud, An overview of geometrical parameters of surface texturing for piston/cylinder assembly and mechanical seals, Meccanica, 2016, 51(1), 9–23.
16. S. Rani, M. Joy and K. P. Nair, Ind. Crops Prod., 2015, 65, 328–333.
17. A. Aravind, M. Joy and K. P. Nair, Ind. Crops Prod., 2015, 74, 14–19.
18. M. Naik, L. C. Meher, S. N. Naik and L. M. Das, Biomass Bioenergy, 2008, 32, 354–357.
19. L. Meher, S. Naik and L. Das, J. Sci. Ind. Res., 2004, 63, 913–918.
20. A. K. Agarwal and K. Rajamanoharan, Appl. Energy, 2009, 86, 106–112.
21. V. Lakshmikanthan, Directorate of Nonedible Oils and Soap Industry, Khadi and Village Industries Commission, Mumbai, India, 1978, vol. 10.
22. S. K. Karmee and A. Chadha, Bioresour. Technol., 2005, 96, 1425–1429.
23. P. T. Scott, L. Pregelj, N. Chen, J. S. Hadler, M. A. Djordjevic and P. M. Gresshoff, BioEnergy Res., 2008, 1, 2–11.
24. N. Jayadas and K. P. Nair, Tribol. Int., 2006, 39, 873–878.
25. S. Asadauskas, J. M. Perez and J. L. Duda, Lubr. Eng., 1996, 52, 877–882.
26. R. Ebewele, A. Iyayi and F. Hymore, Int. J. Phys. Sci., 2010, 5, 841–846.
27. V. M. Devi, P. N. Prasad, L. A. M. Syndia, M. Rajakohila and V. Ariharan, Parameters, 2012, 10, B100.
28. I. R. Fattah, H. Masjuki, M. Kalam, M. Hazrat, B. Masum, S. Imtenan and A. Ashraful, Renewable Sustainable Energy Rev., 2014, 30, 356–370.
29. H.-S. Hwang, A. Adhvaryu and S. Z. Erhan, J. Am. Oil Chem. Soc., 2003, 80, 811–815.
30. B. K. Sharma, A. Adhvaryu, Z. Liu and S. Z. Erhan, J. Am. Oil Chem. Soc., 2006, 83, 129–136.
31. F. Anwar, M. Ashraf and M. I. Bhanger, J. Am. Oil Chem. Soc., 2005, 82, 45–51.
32. H. Imahara, E. Minami and S. Saka, Fuel, 2006, 85, 1666–1670.
33. B. K. Sharma, U. Rashid, F. Anwar and S. Z. Erhan, J. Therm. Anal. Calorim., 2009, 96, 999–1008.
34. M. Kalam, H. Masjuki, M. Shahabuddin and M. Mofijur, Energy, Educ., Sci. Technol., 2012, 30, 123–136.
35. S. Asadauskas, J. H. Perez and J. L. Duda, Tribol. Lubr. Technol., 1997, 53, 35.
36. R. Serrato, M. Maru and L. Padovese, Tribol. Int., 2007, 40, 1270–1275.
37. D. Ahmed, S. Kasolang, R. Dwyer-Joyce, K. Sainan and N. N. Roselina, J. Tribol., 2014, 3, 1–10.
38. Z. Jiang, J. Hutchinson and C. Imrie, Fuel, 2001, 80, 367–371.
39. L. A. Quinchia, M. A. Delgado, J. M. Franco, H. A. Spikes and C. Gallegos, Ind. Crops Prod., 2012, 37, 383–388.
40. M. Noordin and L. Chung, J. Therm. Anal. Calorim., 2009, 95, 891–894.
41. M. M. Conceição, R. A. Candeia, F. C. Silva, A. F. Bezerra, V. J. Fernandes and A. G. Souza, Renewable Sustainable Energy Rev., 2007, 11, 964–975.
42. E. Chiavaro, M. T. Rodriguez-Estrada, C. Barnaba, E. Vittadini, L. Cerretani and A. Bendini, Anal. Chim. Acta, 2008, 625, 215–226.
43. L. Quinchia, M. Delgado, J. Franco, H. Spikes and C. Gallegos, Ind. Crops Prod., 2012, 37, 383–388.
44. C. Garcia, P. Franco, T. Zuppa, N. Antoniosi Filho and M. Leles, J. Therm. Anal. Calorim., 2007, 87, 645–648.
45. J. Santos, L. Lima, I. Santos and A. Souza, J. Therm. Anal. Calorim., 2007, 87, 639–643.
46. V. B. Borugadda and V. V. Goud, Thermochim. Acta, 2014, 577, 33–40.
47. P. van Broekhuizen, D. Theodori, K. Le Blansch and S. Ullmer, Lubrication in Inland and Coastal Water Activities, CRC Press, 2003.
48. M. Garcia-Perez, T. T. Adams, J. W. Goodrum, K. Das and D. P. Geller, Bioresour. Technol., 2010, 101, 6219–6224.
49. S. Z. Erhan, B. K. Sharma, Z. Liu and A. Adhvaryu, J. Agric. Food Chem., 2008, 56, 8919–8925.
50. M. H. Mosarof, M. A. Kalam, H. H. Masjuki, A. Alabdulkarem, A. M. Ashraful, A. Arslan, H. K. Rashedul and I. M. Monirul, Energy Convers. Manage., 2016, 118, 119–134.
51. S. C. Cermak, G. Biresaw, T. A. Isbell, R. L. Evangelista, S. F. Vaughn and R. Murray, Ind. Crops Prod., 2013, 44, 232–239.
52. M. Habibullah, H. H. Masjuki, M. A. Kalam, N. W. M. Zulkifli, B. M. Masum, A. Arslan and M. Gulzar, Ind. Crops Prod., 2015, 76, 188–197.
53. N. Fox, B. Tyrer and G. Stachowiak, Tribol. Lett., 2004, 16, 275–281.
54. M. Shahabuddin, H. H. Masjuki and M. A. Kalam, Procedia Eng., 2013, 56, 597–606.
55. N. Al-Araji and H. Sarhan, Int. J. Eng., 2011, 5, 176–184.
56. L. Havet, J. Blouet, F. R. Valloire, E. Brasseur and D. Slomka, Wear, 2001, 248, 140–146.
57. A. Thanonkaew, S. Wongyai, D. J. McClements and E. A. Decker, LWT--Food Sci. Technol., 2012, 48, 231–236.
58. C. J. Reeves, P. L. Menezes, T.-C. Jen and M. R. Lovell, Tribol. Int., 2015, 90, 123–134.
59. M. Shahabuddin, H. H. Masjuki, M. A. Kalam, M. M. K. Bhuiya and H. Mehat, Ind. Crops Prod., 2013, 47, 323–330.
60. M. H. Mosarof, M. A. Kalam, H. H. Masjuki, A. Alabdulkarem, M. Habibullah, A. Arslan and I. M. Monirul, Ind. Crops Prod., 2016, 83, 470–483.
61. X. Lu, J. Cotter and D. Eadie, Wear, 2005, 259, 1262–1269.
62. Y. Wu, W. Li, M. Zhang and X. Wang, Tribol. Int., 2013, 64, 16–23.
63. N. Jayadas and K. P. Nair, J. Tribol., 2006, 128, 654–659.
64. J.-Q. Hu, X.-Y. Wei, G.-L. Dai, Y.-W. Fei, F. Xie and Z.-M. Zong, Tribol. Int., 2008, 41, 549–555.
65. X. Zeng, H. Wu, H. Yi and T. Ren, Wear, 2007, 262, 718–726.

---