COMPARISON OF VISCOSITY CHARACTERISTICS OF SOYBEAN OILS WITH A MINERAL OIL TWO–STROKE ENGINE LUBRICANT S. Fernando, M. Hanna
ABSTRACT. Soybean oil is a multipurpose oil that is used extensively in industry. The potential use of soybean oil as an alternative lubricant in two–stroke engines was studied. Physical properties of crude, degummed, and methyl ester forms of soybean oil were measured as a function of temperature in preliminary studies. Subsequent laboratory studies of the viscosity of crude and degummed soybean oil revealed that crude and degummed soybean oils had higher viscosities than that of the mineral oil lubricant at a temperature of 1605 C. The viscosity index values of all the soybean oil forms were higher than that of the mineral oil lubricant. Keywords. Soybean oil, Lubricants, Viscosity, Viscosity index, Two–stroke engines, Two–stroke oil.
T
he primary purpose of lubrication is to separate two surfaces sliding past each other with a film of some material that can be sheared without causing damage to the surfaces (Cameron, 1966). As secondary functions, lubricating oils have to cool, clean, and seal engine components (Drake, 1981). The primary raw material used extensively in manufacturing commercial lubricants is crude petroleum oil (Pugh and Court, 1949). Lubricants have also been manufactured from naphthalene derivatives and paraphinic derivatives of mineral oils (Cameron, 1966). In recent years, little information has been published on successful experiences with animal fats and vegetable oils in the lubricant industry. Cameron (1966) and Hersey (1966) cite the earliest lubrication research. They state that the first studies of a lubricated shaft and bearing running under fully hydrodynamic conditions were by Van Pauli in 1849 and by Hirn in 1854, who have been referred to as the fathers of lubrication. Petroff analyzed Hirn’s work in 1883. In 1883 and 1885, Tower also conducted lubrication studies, and Reynolds mathematically analyzed these results in 1886. Reynolds’ paper was the foundation on which all subsequent lubrication theory was based. For over a century, lubricants derived primarily from mineral oils or petroleum distillates have been used to lubricate internal combustion engines. Because mineral oil reserves are non–renewable, and in an effort to address the problem of environmental pollution caused by the use of these petroleum derivatives, attempts have been made to find renewable alternatives, like vegetable oils, to replace mineral
Article was submitted for review in December 2000; approved for publication by the Power & Machinery Division of ASAE in July 2001. The authors are Sandun Fernando, ASAE Student Member, and Milford Hanna, ASAE Fellow Engineer, Professor, Biological Systems Engineering, University of Nebraska, Lincoln, Nebraska. Corresponding author: Milford Hanna, Biological Systems Engineering, University of Nebraska, 211 L.W. Chase Hall, Lincoln, NE 68583–0730; phone: 402–472–1634; fax: 402–472–6338; e–mail:
[email protected].
oil lubricants. Soybean oil has potential as an alternative since soybean oil currently is being used successfully as a lubricant in various components of irrigation systems. In addition, although there is no extensive data on the use of soybean oil as a lubricant, analysis of its physical properties indicates that there is good potential for using soybean oil as a lubricant in internal combustion engines. Finally, better yields under intense cropping systems have drastically reduced soybean prices over the past few years because demand has not expanded relative to yield growth. Therefore, this study was an effort to increase the number of products produced from soybeans. Green (1967) pointed out that, although there are many laboratory tests designed to test the effectiveness of lubricating oils, actual performance can only be discovered after the oils have been used in machines under operating conditions over a period of time. The following is a list of essential requirements that Green suggested a lubricating oil should possess: 1. The oil must be capable of maintaining an efficient lubricating film between all pairs of working surfaces in the engine under operating conditions. 2. The oil must be chemically stable, anti–corrosive, and show good chemical resistance to oxidation in the working environment and through the temperature range over which the engine will operate. 3. The oil should have a high viscosity index, combining easy cold starts and low oil shearing losses with adequate viscosity at maximum running temperature. 4. The oil should preferably have detergent properties capable of inhibiting deposit formation in the engine over the range of operating conditions. 5. The oil must have film strength adequate for bearing surfaces that have very high loadings. 6. For the lubrication of some two–stroke engines, the oil must be miscible with gasoline. We decided to investigate whether soybean oil meets the above requirements. Research has shown that soybean oil has good lubricating properties. The lubricity (also known as film
Transactions of the ASAE Vol. 44(6): 1403–1407
E 2001 American Society of Agricultural Engineers ISSN 0001–2351
1403
strength) values indicate that the lubricating ability of soybean oil is as good as that of mineral oil. During preliminary studies, the miscibility with gasoline, lubricity values, and viscosity values of different soybean oil forms were investigated. The results of these studies paved the way to using soybean oil in two–stroke gasoline engines. The reasons behind this decision were that two–stroke engine lubricating oils do not require commercial additives, and two–stroke gasoline engines run at relatively low temperatures in comparison to diesel and four–stroke gasoline engines. Viscosity is frequently considered the property that determines the effectiveness of any lubricating oil (Georgi, 1950; Green, 1967). The fact that the SAE classification system is based solely on the viscosity of lubricants further supports this statement. However, Green (1967) stated that, because other factors of oil character and quality have not been taken into account, different oils within one SAE viscosity number might vary considerably under operating conditions. As the SAE Handbook (1990) states, changes in viscosity can have marked effects upon an oil’s suitability for certain applications. The ability of an oil to resist changes in viscosity due to changes in temperature is expressed as the viscosity index (V.I.). The viscosity index is an empirical, unitless number. The higher the V.I. of an oil, the less its viscosity changes with temperature. In determining the type of engine that was used in this study, emphasis was given to the operating temperature. As Cameron (1966) pointed out, the operating temperatures in two–stroke engines are generally around 120°C under normal conditions. In addition, as Drake (1981) pointed out, additives like detergents are unnecessary in two–stroke engines. Drake also pointed out that most two–stroke engine manufacturers recommend medium–viscosity lubricants (SAE 20, 30, or 40 oils), which are comparable with the viscosity of soybean oil. For the preliminary studies, we decided to use a manufacturer–recommended lubricant as the control for this experiment. Consequently, in subsequent studies, we decided to use a two–stroke lawnmower from a manufacturer that recommended the control lubricant.
OBJECTIVES A study was designed to address the following objectives with the intention of assessing the suitability of soybean oil as a lubricant for two–stroke internal combustion engines: 1. Compare the miscibility of different soybean oil forms with gasoline. 2. Investigate the viscosity variation of different soybean oil forms in order to assess the suitability of a soybean oil form as an alternative lubricant for two–stroke engines. 3. Determine and compare the viscosity index values of different soybean oil forms with standard two–stroke engine lubricants.
MATERIALS AND METHODS MISCIBILITY The miscibility of different forms of soybean oil with gasoline was studied. The different forms of soybean oil that were compared were crude soybean oil, which was
1404
mechanically expelled; degummed soybean oil, which was solvent extracted; and soybean methyl ester. These soybean oil samples were mixed with gasoline in 1:20 ratios in transparent glass containers to compare miscibilities. The specific proportion was selected because most two–stroke engine manufacturers recommend an oil–to–gasoline ratio in the range of 1:20 to 1:100 (Drake, 1981). Therefore, it was decided to compare and collate the miscibilities at the highest oil concentration. As the first control, two–stroke engine oil was mixed with gasoline at the same ratio. A container of uncontaminated gasoline was utilized as the second control to contrast between the colors of the mixtures. VISCOSITY An experiment was designed to find the absolute viscosity variations with temperature of the three forms of soybean oil and the manufacturer–recommended lubricant. Preliminary test data were used to obtain the expected variance of the new experiment. Analysis showed that three replications were required to obtain satisfactory power for the experiment (Kuehl, 2000). The statistical analysis was performed with statistical software: The SAS System for Windows, release 8.00, and Microsoft Excel 2000. Various statistical procedures were used to analyze the data. The Generalized Linear Models (GLM) procedure was used to test the effect of temperature on viscosity for different oils. Means were compared by Dunnett’s procedure (Littell et al., 1996). The regression between viscosity and temperature was established using Microsoft Excel 2000. The significance of differences, where mentioned, refers to the 95% level (α = 0.05). TREATMENT DESIGN There were four treatments: (1) manufacturer– recommended two–stroke engine oil produced by Toro, (2) soybean methyl ester produced by Soygold, (3) crude soybean oil, a mechanically expelled product of Bruning Grain and Feed Manufacturers, and (4) degummed soybean oil produced by solvent extraction by ADM Inc. Each treatment consisted of three replications, and thus the design included 12 experimental units. Each experimental unit was observed for viscosity at temperature levels ranging from 20°C to 160°C at 10°C intervals. The manufacturer– recommended oil served as the control treatment because this lubricant could be compared with the experimental soybean oil forms for lubricant effectiveness. EXPERIMENT DESIGN A completely randomized design (CRD) was used for the experiment. The experimental units were randomly assigned to the four treatments. The instrument used to find the dynamic viscosity was a Brookfield model DV–E digital viscometer. Because it was necessary for comparison to obtain viscosity values over a range of temperatures, the viscometer was set up to acquire the viscosity values at different temperatures. For heating, a hotplate with a temperature controller was used. The oil container was placed in an oil bath on the hot plate, and the viscosity readings were obtained at temperatures of 20°C to 160°C at 10°C intervals. Viscosity was measured after the sample oil temperature reached the
TRANSACTIONS OF THE ASAE
preset temperature. The temperature readings were measured using a high–temperature glass thermometer. To obtain viscosity values at colder temperatures, the oil samples were cooled, and then the viscosity values were measured while the oil samples warmed up to room temperature. As a control, the viscosity variation of a manufacturer– recommended two–stroke engine oil was measured in the identical manner. Mathematical models were developed to determine the viscosity values at specific temperatures for all the oil forms. Because kinematic viscosity is the standard practice for reporting a lubricant’s viscosity, an experiment was conducted to find the specific gravity of the four oil forms by using the specific gravity bottle. Kinematic viscosity values were then calculated by dividing the dynamic viscosity by the mass density of the oils. VISCOSITY INDEX The viscosity index values were calculated using ASTM standard D 2270–93 (re–approved 1998), adhering to the standard practice for calculating viscosity index from kinematic viscosity at 40°C and 100°C using the following equations (SAE Handbook, 1990). For oils of viscosity index up to and including 100: (1)
where VI = viscosity index L = kinematic viscosity at 40°C of an oil of viscosity index 0 having the same kinematic viscosity at 100°C as the oil whose viscosity index is to be calculated (cSt) H = kinematic viscosity at 40°C of an oil of viscosity index 100 having the same kinematic viscosity at 100°C as the oil whose viscosity index is to be calculated (cSt) U = kinematic viscosity at 40°C of the oil whose viscosity index is to be calculated (cSt). For oils of viscosity index 100 and greater:
where N = (log H – log U)/log Y, or YN = H/U
(2) –6
VI = +100 0.00715
250
(3)
where Y = kinematic viscosity at 100°C of the oil whose kinematic viscosity is to be calculated, mm2/s (cSt).
200
soy methyl ester
150
2
(antilogN )−1
VISCOSITY Figure 1 compares the viscosity variations with temperature of the different soybean oil types and the commercially available two–stroke engine oil (Toro 50:1 two–stroke engine oil). The viscosity variation of the manufacturer– recommended two–stroke engine oil with temperature had a functional relationship of y = 67889x–1.8988 × 10–6, where x is the temperature in °C and y is the kinematic viscosity in m2/s. The viscosity variation of soybean methyl ester with temperature had a functional relationship of y = 40.959x–0.5223 × 10–6, where x is the temperature in °C and y is the kinematic viscosity in m2/s × 10–6. The viscosity variation of crude soybean oil with temperature had a functional relationship of y = 2742x–1.2337 × 10–6, where x is the temperature in °C and y is the kinematic viscosity in m2/s × 10–6. The viscosity variation of degummed soybean oil with temperature had a functional relationship of y = 2602.7x–1.2257 × 10–6, where x is the temperature in °C and y is the kinematic viscosity in m2/s. The R2 values of the above equations are given in table 1.
Viscosity (m /s) x 10
L −U VI = × 100 L−H
soybean oil except less mixing was required. The soybean methyl ester had superlative miscibility properties. The manufacturer–recommended two–stroke engine oil percolated down through gasoline without any sign of mixing. It only mixed with gasoline after vigorous shaking. The degrees of miscibility of all forms of soybean oil were superior to that of the two–stroke engine oil. In the miscibility study, an assumption was made that the same relative viscosity relationships between pure soybean oils and mineral oils would be maintained when both types of oils were mixed with gasoline at the recommended ratios for two–stroke engines. In addition, the above methodology was used because a more standard method was not available to study the miscibility relationships of lubricants and gasoline.
recommended oil crude soybean oil 100
degummed soybean oil best fit curves 50
RESULTS AND DISCUSSION MISCIBILITY When crude soybean oil was poured into a glass container with gasoline, the oil initially percolated to the bottom of the container and then gradually dispersed into the gasoline. The diffusion was clearly visible because the crude soybean oil was yellowish in color. After shaking the contents vigorously for approximately 30 s, the oil and gasoline were completely mixed. The miscibility characteristics of the degummed soybean oil and soybean methyl ester were similar to that of the crude
Vol. 44(6): 1403–1407
0 0
20
40
60
80
100
120
140
160
180
Temperature (o C)
Figure 1. Viscosity variations of different oil forms with temperature. Table 1. R2 values of the best–fit curves for different oil forms. Oil form R2 Manufacturer–recommended oil Soybean methyl ester Crude soybean oil Degummed soybean oil
0.98 0.94 0.99 0.99
1405
Because this study was primarily directed to find alternative two–stroke engine oils, it was necessary to compare the physical properties of the soybean oils with a manufacturer–recommended oil. Table 2 illustrates the comparison of the mean viscosity values at three selected temperatures. Dunnett’s mean comparison procedure was used to compare the means at the 95% significance level. Dunnett’s procedure was used because it effectively controls the type I experimentwise error for comparison of all treatments against a control, unlike the protected LSD, which controls the type I comparisonwise error rate, not the experimentwise error rate. As shown in table 2, all the soybean oil treatment means were significantly different from the control with a negative difference, and the values were greater than 40 × 10–6 m2/s at 40°C. In addition, as shown in figure 1, the manufacturer– recommended oil had considerably higher viscosity values at lower temperatures. At 100°C, there was still a significant difference between all the treatments and the control. However, the differences were reduced drastically from those at 40°C. It is interesting to note that there was no significant difference between treatments 3 and 4 and the control at 160°C. In fact, the trend at the low temperatures was reversed, with the viscosity values of the crude and degummed soybean oil being higher than that of the control. At lower temperatures, the crude and degummed soybean oils had lower viscosity values than the recommended oil, as shown in figure 1. On the other hand, at higher temperatures, the viscosity values of the soybean oils were comparatively high. This suggests that crude and degummed soybean oils have higher viscosity index values. As Drake (1981) pointed out, most two–stroke engine manufacturers recommend lubricating oils classified as SAE 20 or SAE 30. When comparing the physical properties of different forms of soybean oil with SAE 20, it is evident that both crude soybean oil and degummed soybean oil compare well with SAE 20 and ISO 68 VG. One advantage of crude and degummed soybean oils is that they have comparatively higher smoke and flash points than mineral oil
derivatives. This implies that crude and degummed soybean oils have naturally high anti–oxidation properties. VISCOSITY INDEX Figure 2 illustrates the specific gravity variation of different oils with temperature. The given relationships were used to obtain the specific gravity values, which were used for calculating the kinematic viscosity values from the absolute viscosity values. Table 3 lists the viscosity index values that were calculated using the procedure in ASTM standard D 2270. It is interesting to note that all forms of soybean oil have superior viscosity index values in comparison to the manufacturer–recommended oil. As ASTM standard D 2270 points out, the viscosity index is widely used and is an accepted measure of the variation in kinematic viscosity of oil due to changes in temperature. A higher viscosity index (V.I.) indicates a smaller decrease in kinematic viscosity with increasing temperature of the lubricant. Therefore, for the three oils with similar kinematic viscosity values at 160°C, the two soybean oil forms (i.e., crude soybean oil and degummed soybean oil) have higher V.I. values in comparison to the manufacturer–recommended oil. This implies that crude soybean oil and degummed soybean oil have better viscosity–holding properties than the manufacturer–recommended oil, especially at crucial high temperatures. Although the V.I. of soybean methyl ester is the highest of all the oils compared, the viscosity of the soybean methyl ester was comparatively low at low temperatures, and the smoke point was reached at 127°C. Consequently, the use of soybean methyl ester as a lubricant in internal combustion engines is precluded when the operating temperatures in the engine block are higher than 127°C. The tests of miscibility, viscosity, and viscosity index suggest that crude and degummed soybean oils have a high propensity for succeeding as alternative lubricating oils in 0.94 recommended 2–S oil y = –0.0006x + 0.9358
0.92
soybean oil
0.9
Table 2. Mean comparisons of viscosity values at different temperatures using Dunnett’s mean comparison procedure. Difference Simultaneous 95% between confidence limits means of the difference Treatment 2 –6 Temperature (m /s × 10 ) between means comparison 160_C
100_C
40_C
[a]
0.88
soy methyl ester
0.86 y = –0.0005x + 0.8833
0.84 0.82
3 – 1[a] 4–1 2–1
0.45 0.37 –2.35
–2.62 –0.10 –2.62
0.86 0.64 –1.87 [b]
3–1 4–1 2–1
–1.04 –1.27 –5.98
–1.94 –2.27 –6.88
–0.14 [b] –0.37 [b] –5.08 [b]
Figure 2. Specific gravity variation of different oil soybean oil forms and a commercial two–stroke engine oil with temperature.
3–1 4–1 2–1
–39.62 –40.75 –65.17
–63.71 –64.84 –89.28
–15.53 [b] –16.66 [b] –41.10 [b]
Table 3. Viscosity index values calculated using ASTM standards. Oil type Viscosity index
Treatment 1 = manufacturer–recommended oil Treatment 2 = soybean methyl ester Treatment 3 = crude soybean oil Treatment 4 = degummed soybean oil. [b] Comparisons significant at the 0.05 level
1406
y = –0.0007x + 0.8985
0.8 0.78 0
50
100
150
200
o
Temperature C
Manufacturer–recommended oil Soybean methyl ester Crude soybean oil Degummed soybean oil
116.45 853.43 273.46 275.35
TRANSACTIONS OF THE ASAE
two–stroke engines. Nevertheless, there have been concerns of carbon deposits when using untransesterfied oils from renewable resources in diesel engines. Subsequent two–stroke engine testing, which is beyond the scope of this article, revealed that there was little evidence of such deposits at proper oil–to–gasoline mixtures. Our interpretation was that carbon deposits would be minimized with the low level of oil substitution used in this study. Longer–term implications will have to be documented.
2. More extensive engine tests should be conducted to determine the optimum ratio of crude soybean oil to gasoline for commercial usage. 3. The prospective soybean oil and derivatives should be tested in a demonstration study with different types of two–stroke machinery under different load conditions. 4. Different vegetable oil types should be blended with soybean oil to improve the physical properties of the end product for better performance and wider application.
CONCLUSIONS AND RECOMMENDATIONS
REFERENCES
The miscibility of soybean oil with gasoline is satisfactory when mixed in an oil–to–gasoline ratio of 1:20 at a temperature of 25°C. The viscosity values of crude and degummed forms of soybean oil fall within the viscosity standards of SAE 20, or thinner, lubricants. The viscosity of crude soybean oil is close to SAE 30. The viscosity index values of the soybean oil forms are higher than that of the manufacturer–recommended two–stroke engine oil. Consequently, the effect of temperature was greater on the viscosity of manufacturer–recommended oil than on the soybean oil forms. Therefore, it could be concluded that the viscosity characteristics of crude and degummed soybean oils at higher temperatures were equal or superior to those of the recommended petroleum derivatives. In order to move ahead with utilizing soybean oil as a commercial two–stroke engine lubricant, further studies should be performed under laboratory and field conditions. The proposed studies are: 1. Soybean oil and derivatives should be subjected to standard lubrication tests.
Cameron, A. 1966. The Principles of Lubrication. New York, N.Y.: John Wiley and Sons. Drake, G. R. 1981. Small Gasoline Engines: Maintenance, Troubleshooting, and Repair. Reston, Va.: Reston Publishing Co. Georgi, C. W. 1950. Motor Oils and Engine Lubrication. New York, N.Y.: Reinhold Publishing. Green, A. B. 1967. Lubrication and Lubricants. E. R. Brainthwaite, ed. New York, N.Y.: Elsevier Publishing Co. Hersey, M. D. 1966. Theory and Research in Lubrication: Foundation for Future Developments. New York, N.Y.: John Wiley and Sons. Kuehl, R. O. 2000. Design of Experiments: Statistical Principles of Research Design and Analysis. 2nd ed. Belmont, Cal.: Brooks/Cole Inc. Littell, R., C. Milliken, G. A. Stroup, and R. D. Wolfinger. 1996. SAS system for mixed models. Cary, N.C.: SAS Institute Inc. Pugh, B., and J. M. A. Court. 1949. Fuels and Lubricating Oils for Internal Combustion Engines. London, U.K.: Sir Isaac Pitman and Sons Ltd. SAE Handbook. 1990. Vol. 03: Engines, fuels, lubricants, emissions, and noise. Warrendale, Pa.: Society of Automotive Engineers, Inc.
Vol. 44(6): 1403–1407
1407
1408
TRANSACTIONS OF THE ASAE
