Article pubs.acs.org/JAFC
Synthesis and Tribological Investigation of Lipoyl Glycerides Girma Biresaw,*,† Joseph A. Laszlo,‡ Kervin O. Evans,‡ David L. Compton,‡ and Grigor B. Bantchev† †
Bio-Oils Research Unit and ‡Renewable Product Technology Research Unit, National Center for Agricultural Utilization Research, Agricultural Research Service, U.S. Department of Agriculture, 1815 North University Street, Peoria, Illinois 61604, United States ABSTRACT: Lipoyl glycerides were synthesized by enzymatic transesterification of lipoic acid with high-oleic sunflower oil in 2methyl-2-butanol solvent. The synthesis gave a crude product mixture comprising unreacted lipoic acid, free fatty acids, and several lipoyl glyceride structures of varying lipoic acid substitution. A more purified product mixture, devoid of unreacted lipoic acid and free fatty acids, was obtained in 61% yield. The crude and purified product mixtures were thoroughly characterized and their components positively identified. The tribological properties of the product mixtures were further investigated using a variety of methods. The product mixtures displayed significantly improved oxidation stability, cold-flow, and extreme pressure properties over those of the parent high-oleic sunflower oil. The extreme pressure results for the neat products showed a higher weld point for the crude than for the purified mixture. This was attributed to differences in the chemical properties of the components in the two product mixtures. KEYWORDS: 4-ball antiwear, 4-ball extreme pressure, biolubricant, cold-flow properties, enzymatic transesterification, high-oleic sunflower oil, lipoic acid, oxidation stability, soybean oil, viscosity, viscosity index
■
INTRODUCTION Currently, most lubricants on the market are formulated using ingredients derived from petroleum-based rather than biobased raw materials.1,2 This is mainly because of the perception that petroleum-based products are cheaper than biobased products. However, careful scrutiny of the overall cost of lubrication indicates that biobased lubricants, derived from vegetable oils or other renewable materials, provide significant cost advantages over the life of the lubricant.3 Vegetable oils have a number of favorable characteristics that make them attractive raw materials for use in lubricant formulations.1−4 These characteristics provide vegetable oils with superior economic, environmental, and performance advantages over petroleumbased raw materials.1−4 Vegetable oils are derived from renewable and sustainable farm products as opposed to petroleum, which is rapidly depleting. The use of vegetable oils in lubrication and other industrial products provides new markets to surplus agricultural crops and thereby strengthens the economic well-being of rural and farm communities. Because vegetable oils are used for human consumption and animal feed, their use in lubrication and other industrial products provides safer alternatives to petroleum-based products. As a result, lubricants formulated from vegetable oil derived ingredients are environmentally friendly and safe during manufacture and use and when disposed after use. Vegetable oils also possess a number of tribological properties that are superior to those of petroleum-based ingredients. These include excellent viscosity index, very low volatility, very high biodegradability, very high compatibility with and enhancement of extreme pressure additives, and low traction coefficient.4−8 However, despite the above-listed advantages, vegetable oils do have some inherent and serious weaknesses relative to petroleum-based ingredients used in lubricant formulations. Among the problems commonly encountered are poor oxidation stability, high pour point and This article not subject to U.S. Copyright. Published 2014 by the American Chemical Society
cloud point temperatures, poor hydraulic stability, and poor bioresistance. Over the years, various approaches have been applied to mitigate these weaknesses. The simplest practice is that of blending vegetable oils with petroleum-based oils, such as polyalphaolefins, that have superior oxidation stability or other desirable properties.4,9 Another widely used practice is blending commercial additives, such as antioxidants and pour point depressants, to improve specific properties.9,10 A third approach involves synthetically replacing the structure of the vegetable oil causing the weakness. Chemical, thermal, and enzymatic synthetic methods have been applied in this approach.11−14 An example of such an approach is to synthetically convert unsaturations in the vegetable oils to other functional groups to improve oxidation stability. Poor oxidation stability of vegetable oils is associated with the high reactivity of allylic and bis-allylic protons.15 Conversion of the double bonds to epoxides or other functional groups eliminates allylic and bis-allylic protons and thereby improves the oxidation stability of vegetable oils.16 Synthesis can also be used to convert vegetable oils into multifunctional lubricant ingredients. Such ingredients perform several functions in the lubricant that are normally performed by several specific additive ingredients. By doing so, multifunctional ingredients reduce the complexity of lubricant formulations because it allows for reducing the number of components in the formulation. As a result, the formulation will be easier to analyze and control. Conversion of vegetable oils to multifunctional ingredients also increases the value of vegetable oils for lubricant formulation applications. Normally, vegetable oils are used as base oils in lubricant formulations. Synthetic conversion to multifunctional ingreReceived: Revised: Accepted: Published: 2233
September 30, 2013 February 13, 2014 February 18, 2014 February 27, 2014 dx.doi.org/10.1021/jf404289r | J. Agric. Food Chem. 2014, 62, 2233−2243
Journal of Agricultural and Food Chemistry
Article
tuning process. The column (Luna Phenyl-Hexyl; 5 μm, 250 mm × 4.6 mm) was eluted isocratically with 95% acetonitrile/5% acetone containing 0.1% formic acid at a flow rate of 1.0 mL/min, with UV detection at 340 nm. Solubility in Petroleum-Based Oils. The solubility of lipoyl glyceride product mixtures in 4 at room temperature was determined gravimetrically, using visual inspection for any changes in oil transparency, and is expressed as a percentage of the total weight. Surface Tension. Equilibrium surface tension at room temperature (23 ± 2 °C) was measured using the axisymmetric drop shape analysis method17 on the FTA 200 automated goniometer (First Ten Angstroms, Portsmouth, VA, USA), using previously reported procedure.18 Density, Viscosity, and Viscosity Index. Density and dynamic and kinematic viscosity as a function of temperature were measured on a Stabinger SVM3000/G2 viscometer (Anton Paar GmbH, Graz, Austria). Viscosity index was calculated from kinematic viscosity data at 40 and 100 °C following the procedure outlined in ASTM D 227093.19 Pressurized Differential Scanning Calorimetry (PDSC). PDSC tests were conducted on a Q20P pressure differential scanning calorimeter (TA Instruments−Waters LLC, New Castle, DE, USA) fitted with a computer and appropriate software to allow for data acquisition and analysis. All tests were conducted with the cell pressurized with pure oxygen to 500 ± 25 psig in dynamic mode, that is, with a positive oxygen flow rate of 100 ± 10 mL/min. In a typical experiment, an oil sample of 1.2−2.0 mg was placed onto an aluminum pan, which was then hermetically sealed; a small pinhole was punched on the top cover, and the pan was placed in the instrument cell. The cell was then pressurized to 500 ± 25 psig and the temperature allowed to equilibrate to 50 °C. Oxygen was then allowed to flow at 100 ± 10 mL/min. Once the desired oxygen pressure, flow rate, and initial cell temperature were attained, the cell temperature was ramped up to 250 °C at a rate of 10 °C/min. The resulting heat flow versus temperature data were plotted and analyzed using the instrument computer and software to determine onset temperature and peak temperature. Duplicate runs were conducted, and average values are reported. Four-Ball (4-Ball) Instrument and Test Specimen. The 4-ball tests were conducted on a model KTR-30L 4-ball tribometer equipped with TriboDATA software (Koehler Instruments, Bohemia, NY, USA). The specification and detailed description of the instrument hardware and software have been given before.6 Test balls used in 4ball experiments were obtained from Falex Corp. (Aurora, IL, USA), and their specifications have also been given before.6 4-Ball Antiwear Test. The 4-ball antiwear tests were conducted according to the procedure outlined in ASTM D 4172-94.20 The coefficient of friction for each test was calculated from the corresponding torque and load data using the procedure outlined in ASTM D 5183-95.21 The wear scar diameters of the balls used in the test were measured using a wear scar measurement system comprising hardware and ScarView software (Koehler Instruments). The wear scar diameters along and across the wear direction of the three balls were used to calculate the average wear scar diameters for the test. Each test lubricant was used in at least two antiwear measurements, and average coefficient of friction and wear scar diameters values were reported. 4-Ball Extreme Pressure Test. The 4-ball extreme pressure test was conducted according to the procedure outlined in ASTM D 278388.22 The test comprises a series of 10 s tests at increasing loads until welding of the four balls is observed. The load at which welding is observed is the weld point and is a characteristic extreme pressure property of the lubricant tested. For tests below the weld point, the average wear scar diameter of the three balls, measured in accordance with ASTM D 4172-94,27 is recorded as a function of the test load.
dients allows vegetable oils to perform as base oil and as additive(s). The additive function depends on the chemical nature of the group introduced into their structure. For example, introduction of branching through synthetic conversion will allow the vegetable oil to also take on the role of a pour point depressant additive because it will result in lowered pour point and cloud point properties.12,16 Thus, such multifunctional characteristics make it unnecessary to incorporate, analyze, and control pour point depressant in the formulation, which is a big cost savings for the lubricant user. This study was designed to enzymatically synthesize, purify, identify, and characterize crude and purified lipoyl glyceride product mixtures for extreme pressure and other tribological properties.
■
MATERIALS AND METHODS
Materials. α-Lipoic acid (D/L racemic mixture, 1) and 2-methyl-2butanol were purchased from Sigma-Aldrich (St. Louis, MO, USA). Novozym 435 (Candida antarctica lipase B immobilized on acrylic beads) was purchased from Novozymes North America (Franklinton, NC, USA). High-oleic sunflower oil (81% oleic acid, 2) was purchased from Columbus Foods (Des Plaines, IL, USA). Refined, bleached, and deodorized soybean oil, 3, was obtained from Pioneer Hi-Breed International (Des Moines, IA, USA). Polyalphaolefin with a viscosity of 6 cSt at 100 °C (PAO6), known under the trade name Durasyn 166, 4, was a free sample from Ineos Oligomers (League City, TX, USA). Di-tert-dodecyl polysulfide, (t-C12H25)2S5, known under the trade name TPS-32, was a free sample supplied by Arkema Canada Inc. (Burlington, ON, Canada). Synthesis of Lipoyl Glycerides. Lipoic acid, 1 (20.6 g; 0.10 mol), high-oleic sunflower oil, 2 (133 g; 0.15 mol), and 2-methyl-2-butanol (46 g) were treated with Novozym 435 (15.4 g; 10:1 substrate to enzyme ratio) for 24 h at 60 °C in an orbital shaker. The reaction mixture was then placed under partial vacuum (250 mbar) for 6 h at 60 °C using a rotary evaporator, which served to slightly increase the generation of lipoyl glycerides. The enzyme was separated from the product by filtration through a 20 μm nylon Spectra/Mesh (ColeParmer, Vernon Hills, IL, USA) and washed with acetone (200 mL) to increase product recovery. Solvent was removed under reduced pressure at 60 °C, and 133 g (89% yield) of the crude lipoyl glyceride product was obtained. Solvent Fractionation of Lipoyl Glycerides. The reaction product was extracted with methanol (2 × 300 mL) to remove lipoic acid and free fatty acids. The oil phase was freed of residual solvent under reduced pressure at 60 °C to give 93 g, 61% isolated yield, of purified lipoyl glyceride product. HPLC Analysis. Analyses were performed using a Thermo Separation Products (San Jose, CA, USA) HPLC system consisting of a Spectra System AS3000 autosampler, a Spectra System P4000 pump, a Spectra System UV6000LP detector, an Alltech (Deerfield, IL, USA) 500 evaporative light-scattering detector, and a 250 mm × 4.6 mm i.d., 5 μm, Luna Phenyl-Hexyl column (Phenomenex, Torrance, CA, USA). The column was eluted isocratically with 95% acetonitrile/ 5% acetone/0.05% acetic acid (v/v/v) at a flow rate of 1.5 mL/min, with UV detection at 340 nm. Injection volumes were 10 μL. LC High-Resolution Mass Spectroscopy Analysis. Structural confirmation of derivative structures was obtained by high-resolution mass spectrometry using a Thermo Scientific Accela UHPLC system (autoinjector, PDA detector, and a 1250 quaternary pump) coupled with an LTQ Orbitrap Discovery mass spectrometer (MS), with an Ion Max electrospray ionization (ESI) source, all running under Thermo Scientific Xcalibur LC-MS software. The MS was run with the ESI probe in the positive mode. The source inlet temperature was 350 °C; the sheath gas rate was set at 10 arbitrary units; the auxiliary gas rate was set at 2 arbitrary units; and the sweep gas rate was set at 2 arbitrary units. The maximal mass resolution was set at 30000, the spray voltage was set at 3.0 kV, and the tube lens was set at −100 V. Other parameters were determined and set by the calibration and
■
RESULTS AND DISCUSSION Lipoic Acid Basics. Lipoic acid, 1, is a naturally occurring cellular cofactor that has a cyclic disulfide moiety.23 Sulfur is
2234
dx.doi.org/10.1021/jf404289r | J. Agric. Food Chem. 2014, 62, 2233−2243
Journal of Agricultural and Food Chemistry
Article
species. The reaction product was a complex mixture of unreacted starting materials (1 and 2) as well as various structures of lipoyl glycerides and free fatty acids (9). Some of the structures in the reaction product mixture are depicted in Figure 1. HPLC and LC-MS analyses were used to identify and quantitate 1 and the various lipoyl products in the product mixture. The following principal lipoyl reaction products were observed: lipoyl glycerol (5), lipoyl monooleoylglycerol (6), dilipoyl oleoylglycerol (7), and lipoyl dioleoylglycerol (8). The mass spectra of these entities are shown in Figure 2. In all cases, the parent compound was identified in a complex with a metal ion (K+ or Na+) or formate resulting from the MS ionization process. Multiple complexes were observed, including dimers. The fact that the principal products are oleoylglycerides is consistent with the fatty acid composition of 2, which is mainly (81%) oleic acid (Table 1).31 As expected, small quantities of similar lipoyl derivatives of other fatty acids of 2 were also observed. The analysis of the lipoyl product mixture showed that 72 mol % of the original 1 participated in the transesterification reaction to produce a variety of lipoyl products, whereas 28 mol % did not react at all (Table 2). The relative molar composition of reacted 1 in the crude product mixture decreased in the order 8 > 7 > 6 > 5 (Table 2). The crude product mixture containing lipoyl glycerides was extracted with a polar solvent, methanol. The resulting purified mixture was analyzed using the same procedures used to analyze the crude product mixture. Table 3 compares the composition (% w/w and mol/mol) of the major species in the crude versus purified product mixtures. As expected, extraction with methanol completely removed the more polar species from the crude product. As a result, the purified product was free of unreacted 1 and 9 generated by the transesterification reaction. These two species are the most polar components in the crude product mixture. Their removal resulted in an increase in the relative proportion of the less polar components 2, 8, and 7 in the purified product mixture (Table 3). Thus, the crude and purified product mixtures represented two different compositions and were evaluated for their physical and tribological properties. Density. The densities of neat soybean oil (3), 2, crude and purified product mixtures, and blends at 40, 75, and 100 °C are compared in Table 4. As expected, the densities of all the oils investigated decreased with increasing temperature. The data in Table 4 show that the crude and purified product mixtures had higher densities than the starting 2, which had lower density than 3. The higher density of the product mixtures over 2 can be attributed to the insertion of the heavier sulfur atoms during transesterification. The density results also show a slightly higher value for the crude than for the purified product mixture. However, for all practical purposes, the two product mixtures can be considered to have similar densities. The data in Table 4 also show that the blends have higher densities than the neat vegetable oils (2, 3), and the density increases proportionately with increasing concentrations of the product mixtures in 2 and 3. An important observation in Table 4 is the relative density of the crude product mixture blend in 2 versus that in 3. The density of the blend in 3 is greater than that in 2 at 40 and 75 °C but is reversed (2 > 3) at 100 °C. This could be due to a higher propensity of crude product mixture to decompose in 3 than in 2 at 100 °C, which is close to the melting point of sulfur.
one of several elements known to provide extreme pressure properties to lubricants.24 This makes derivatives of lipoic acid potential candidates for application in extreme pressure lubrication. Lipoic acid is an ingredient in dietary supplements and cosmetics.25,26 In addition, lipoic acid exhibits beneficial effects in the therapy of many diseases. In the past few years, several papers have described its chemical modifications to improve pharmacological properties.27 The solubility of lipoic acid in lubricating oils is very low but can be significantly improved with appropriate modifications. Novel conjugates of 1,3-dioleoyl-glycerol and lipoic acid have been already described in the literature.28 The enzymatic route described in this paper is a green synthesis approach that is readily scalable to high production volumes through proven continuous packed-column technology29 and does not rely on single-use catalysts described in the literature28 procedure. Synthesis and Characterization of Lipoyl Glycerides. A schematic for the enzymatic synthesis of lipoyl glycerides is given in Figure 1. A detailed description of the synthesis, purification, and characterization of lipoyl glycerides has been given previously.30 Various glycerides of lipoic acid (1) were obtained by a lipase-catalyzed transesterification reaction between high-oleic sunflower oil (2) and 1 (Figure 1). A 1:1.5 molar ratio of 1 to 2 was used. Application of excess 2 in the reaction helped ensure that a large proportion of 1 was incorporated into the glyceride
Figure 1. Schematic of the transesterification reaction of lipoic acid with high-oleic sunflower oil. 2235
dx.doi.org/10.1021/jf404289r | J. Agric. Food Chem. 2014, 62, 2233−2243
Journal of Agricultural and Food Chemistry
Article
Figure 2. Mass spectra of (A) lipoyl glycerol, (B) lipoyl monooleoylglycerol, (C) dilipoyl oleoylglycerol, and (D) lipoyl dioleoylglycerol. The asterisk in each panel marks the identifying peak for the illustrated compound.
The data in Table 4 also show that the kinematic and dynamic viscosity of blends in 2 or 3 increased with increasing concentrations of product mixtures in the blend. The viscosity indices of neat 2, 3, crude and purified product mixtures, and blends are also given in Table 4. Transesterification resulted in slightly lowering the viscosity index of 2 from about 200 to 178−185. Despite this lowering, the viscosity index range for the neat product mixtures was much higher than the best petroleum-based and polyalphaolefin (4) oils. Blending of lipoyl glyceride product mixtures in 2 and 3 caused a slight reduction in the viscosity index of these two vegetable oils. As shown in Table 4, the viscosity index decreased with increasing concentration of the two product mixtures in the vegetable oils. Surface Tension. The surface tensions of the neat oils and blends are compared in Table 5. As shown in Table 5, within one standard deviation, transesterification did not cause any change in the surface tension of 2. The surface tension of 2, which is slightly higher than that of 3, remained more or less the same. However, as shown in Table 5, blending of the crude product mixtures into 3 increased the surface tension of 3 by almost 1 dyn/cm to the value of the neat product mixtures. Solubility in Polyalphaolefin Oil (4). Application of lipoyl glycerides as lubricant additives requires that they be soluble
Table 1. Fatty Acid Composition of High-Oleic Sunflower (2) and Soybean (3) Oils fatty acid name
fatty acid structure
2, % w/w31
3, % w/w5
palmitic stearic oleic linoleic linolenic
C16:0 C18:0 C18:1 C18:2 C18:3
5 4 81 9 0.5
10.6 4.0 23.2 53.7 7.6
Viscosity and Viscosity Index. The kinematic and dynamic viscosities of neat 2, 3, crude and purified product mixtures, and blends are compared in Table 4. As with density, the viscosities of the transesterification products were higher than that of the starting 2, which was more viscous than 3 (Table 4). The increased viscosity of the two product mixtures relative to 2 can be attributed to the increased polarity of the molecules after transesterification. The increased polarity will increase the intermolecular interactions between the molecules and thereby retard oil flow. Table 4 also shows that the purified product mixture is more viscous than the crude, which is the reverse of the relative densities between these two product mixtures. This could be due to the higher average molecular weight of the purified product mixture relative to the crude. 2236
dx.doi.org/10.1021/jf404289r | J. Agric. Food Chem. 2014, 62, 2233−2243
Journal of Agricultural and Food Chemistry
Article
Table 2. Relative Molar Distribution of Lipoyl Species in the Crude Lipoyl Glyceride Product Obtained from the Transesterification Reaction between Lipoic Acid and High-Oleic Sunflower Oil lipoic moiety distribution,a mol %
lipoic species
28 ± 1
1, lipoic acid
ND
6±1
5, lipoyl glycerol
m/z (found)b
m/z (calcd)
c
d
NDd
325.078 C12H21O6S2
325.078 [M + formate]−
6, lipoyl monooleoylglycerol
12 ± 2
583.289 C29H52O5S2K
583.306 [M + K]+
7, dilipoyl oleoylglycerol
15 ± 1
755.348 C37H64O6S4Na
755.348 [M + Na]+
8, lipoyl dioleoylglycerol
39 ± 2
847.535 C47H84O6S2K
847.532 [M + K]+
a
Mean and standard deviation (n = 3) of the lipoic species (relative to the initial moles of lipoic acid) concentration based on detected UV absorbance (340 nm) during HPLC. Includes oleic and other fatty acid species. bDetermined by LC-MS. c28 mol % of the starting lipoic acid did not participate in the transesterification reaction. dNot determined.
Table 3. Composition of Crude (LGc) and Purified (LGp) Lipoyl Glyceride Product Mixture Obtained from Transesterification of Lipoic Acid with High-Oleic Sunflower Oil %, mol/mol component
mol wt, g/mol
1, lipoic acid 9, free fatty acids 5, lipoyl glycerol 6, lipoyl monooleoylglycerol 7, dilipoyl oleoylglycerolc 8, lipoyl dioleoylglycerol 2, high-oleic sunflower oil LGc (wt av) LGp (wt av) total [S], % w/w
206 282 280 544 733 809 885 701 831
a
%, w/w b
LGc
LGp
10.2 35.4 2.4 5.1 3.0 15.6 28.2
0.0 0.0 2.9 3.0 6.7 41.3 46.1
[S] % in b
LGc
LGp
4.0 18.0 1.3 5.0 4.0 22.8 45.0
0.0 0.0 1.0 2.0 6.0 41.0 50.0
LGc
LGp
1.2 0.0 0.3 0.6 0.7 1.8 0.0
0.0 0.0 0.2 0.2 1.1 3.2 0.0
4.6
4.7
a
Molecular weight calculations are based on oleic acid as the fatty acid in the component. bReaction product after solvent fractionation. cTaking into account that dilipoyl oleoylglycerol has two lipoyl groups, its molecular species portion was considered to be half of its proportion measured by HPLC at 340 nm.
Table 4. Density (d), Kinematic Viscosity (kVis), Dynamic Viscosity (dVis), and Viscosity Index (VI) of Neat Oils (2, 3, and Crude (LGc) and Purified (LGp) Lipoyl Glyceride Product Mixtures) and Blendsa neat oils
blends of LGc in 2
T (°C)
3
2
LGc
LGp
5% w/w
10% w/w
5% w/w
10% w/w
d, g/mL
40 75 100
0.9068 0.8838 0.8671
0.8994 0.8763 0.8683
0.9284 0.9045 0.8902
0.9274 0.9040 0.8875
0.9080 0.8847 0.8684
0.9087 0.8854 0.8690
0.9006 0.8775 0.8711
0.9021 0.8790 0.8761
kVis, mm2/s
40 75 100
31.88 12.57 7.68
40.77 14.69 8.72
42.18 14.44 8.35
45.96 15.74 9.12
33.15 12.72 7.80
33.26 12.69 7.74
40.85 14.635 8.67
40.03 14.36 8.50
dVis, mPa-s
40 75 100
28.91 11.11 6.66
36.67 12.87 7.5
39.25 13.09 7.44
42.62 14.23 8.1
30.10 11.26 6.78
30.22 11.24 6.73
36.42 12.75 7.43
36.67 12.78 7.43
224
200
178.5
185.0
218.5
215.0
199
197.5
VI a
blends of LGc in 3
Standard deviations: d, ±0.0000−0.0199 g/mL; kVis, ±0.00−0.58 mm2/s; dVis, ±0.00−0.81 mPa-s; VI, ±0.00−2.1
results of the solubility study in 4, which has a viscosity close to that of 3, are also summarized in Table 5. As shown in Table 5, the crude product mixture was found to be insoluble in 4,
not only in biobased base oils such as 2 and 3 but also in petroleum-based oils such as 4. Solubility investigation showed that both product mixtures are highly soluble in 2 and 3. The 2237
dx.doi.org/10.1021/jf404289r | J. Agric. Food Chem. 2014, 62, 2233−2243
Journal of Agricultural and Food Chemistry
Article
Table 5. Solubility, Surface Tension, Cloud Point, and Pour Point of Neat Oils (2, 3, and Crude (LGc) and Purified (LGp) Lipoyl Glyceride Product Mixtures) and Some Blends solubility in polyalphaolefin oil, % w/w neat oils 3 2 LGc LGp
surface tension, dyn/cm 31.8 32.7 32.8 33.0
11.0
± ± ± ±
0.5 0.2 0.3 0.3
cloud point, °C
pour point, °C
−3.5 ± 0.3 −1.4 ± 0.2
−95 −129 −17.7 ± 0.6 −16.3 ± 0.6
blends 32.6 ± 0.1 32.7 ± 0.1
5% LGc in 3 10% LGc in 3
Figure 3. PDSC: (A) onset temperature (OT) and (B) peak temperature (PT) of 2, 3, crude (LGc) and purified (LGp) lipoyl glyceride product mixtures, and blends (°C).
whereas the purified product mixture was highly soluble. The difference in the solubility of these two product mixtures is a reflection of the difference in their compositions. As shown in Table 3, the crude contains almost 48% mol/mol of polar or hydrophilic components (1, 5, 9), whereas the purified
contained a mere 2.9% mol/mol. As a result, the crude mixture is hydrophilic and insoluble in the lipophilic 4, whereas the purified mixture is lipophilic and highly soluble in 4, as shown in Table 5. 2238
dx.doi.org/10.1021/jf404289r | J. Agric. Food Chem. 2014, 62, 2233−2243
Journal of Agricultural and Food Chemistry
Article
Figure 4. Antiwear properties of 2, 3, crude (LGc) and purified (LGp) lipoyl glyceride product mixtures, and blends: (A) coefficient of friction (COF); (B) wear scar diameter (WSD, mm).
Oxidation Stability. Oxidation stability was investigated using PDSC. In this procedure, the oils are heated at a fixed rate (°C/min) under a pressurized oxygen atmosphere until oxidation is detected due to changes in heat flow. The temperatures at which the oxidation begins, called the onset temperature, and maximum oxidation occur, called the peak temperature, are used to compare the oxidation stability properties of oils. Onset and peak temperatures correlate very well, and either or both values can be used to compare the oxidation stability of vegetable oils. The higher the onset and peak temperatures, the more stable the oil is toward oxidation. The oxidation of vegetable oils has been attributed to the presence of allylic and bis-allylic protons in their structures.2,15 Thus, the relative abundance of mono- and polyunsaturated fatty acids in the respective triglycerides of the vegetable oils has a strong influence on oxidation stability properties. The higher the degree of unsaturation, the less oxidatively stable the oil. Table 1 compares the fatty acid compositions of 2 and 3 and provides data about the structure of the fatty acid as it relates to the number of double bonds as well as the relative fraction of each fatty acid in the triglyceride mixture in the oil. The data in
Table 1 can be used to estimate the degree of unsaturation or unsaturation index (UI) for the vegetable oils as32 UI = 100 × [Σ(d i × wi)]
(1)
where di is the number of double bonds on a fatty acid and wi is the weight fraction of the fatty acid in the vegetable oil. According to eq 1, the larger the number of double bonds in the fatty acid and the larger its weight fraction among the fatty acids in the triglyceride, the higher the UI value and the less oxidatively stable the oil will be. Application of eq 1 to the data in Table 1 gave UI values of 100.5 and 153.4 for 2 and 3, respectively. Thus, eq 1 predicts that 2 will have a lower UI and, hence, will be more oxidatively stable than 3. Figure 3 compares the PDSC results for neat 2, 3, crude and purified product mixtures, and their blends. The data for the neat vegetable oils shows higher onset and peak temperatures for 2 than for 3, as was predicted by eq 1. The data in Figure 3 also show that the neat crude and purified product mixtures have significantly higher oxidation stability than either vegetable oil. It also shows that the addition of small amounts of the crude or purified product mixture to the vegetable oils resulted in significant improvement in their oxidation stability. The 2239
dx.doi.org/10.1021/jf404289r | J. Agric. Food Chem. 2014, 62, 2233−2243
Journal of Agricultural and Food Chemistry
Article
Figure 5. Weld points (WP) of 2, 3, crude (LGc) and purified (LGp) lipoyl glyceride product mixtures, and blends (kgf).
Two possible reasons are replacement of long-chain fatty acids (C16 and higher) in the starting triglycerides, which causes crystallization and solidification at lower temperature, by the shorter chain (C8) 1 during transesterification (see Figure 1 and Table 1), and production of a much more complex product mixture after transesterification than that of the starting 2, which will result in depressed melting temperatures. Antiwear Properties. 4-Ball friction and wear properties under antiwear conditions (40 kgf, 1200 rpm, 75 °C, 60 min) are compared in Figure 4. The test provides two parameters: coefficient of friction and wear scar diameter. Neat 2 showed a higher coefficient of friction and higher wear scar diameter than neat 3. This result is contrary to expectation based on the relative viscosities of these two vegetable oils (Table 4). Antiwear tests are conducted under mixed film conditions where both the viscosity and chemical properties of the oils will have an effect. In oils with similar chemical properties, viscosity becomes the dominant predictor of antiwear results. In this particular case, however, viscosity effects were overwhelmed by chemical effects. This indicates major differences in chemical properties between these two vegetable oils, far beyond what is quantified by the fatty acid composition shown in Table 1. Comparison of the antiwear results for the two product mixtures showed a different result from that between 2 and 3. Here, the two product mixtures have similar chemical origins. Both are transesterification products of 1 and 2, which means they have similar chemical species in their mixture. However, because of the differences in purification procedures, they comprise different proportions of the same species. The consequence of this, as discussed previously, was differences in lipophilicity and viscosity. The crude mixture was found to be less viscous and less lipophilic than the purified (Tables 3 and 4). As a result, the crude mixture, because of its hydrophilic properties, was found to be insoluble in the lipophilic base oil 4 (Table 5). The consequence of viscosity is illustrated in Figure 4, where the lower viscous crude product displayed a poorer coefficient of friction and wear scar diameter. Such a result is expected in mixed film test conditions of oils with similar chemical properties but different viscosities.24 Extreme Pressure (EP) Properties. Extreme pressure tests are conducted to determine if the oils are capable of providing lubrication at very high pressures and/or temperature
onset and peak temperatures of the vegetable increased with increasing concentration of crude and purified product mixtures in the blend. There are two possible reasons for the improvements in the oxidation stability of 2 after the transesterification reaction with 1. The most obvious reason is the replacement of fatty acids with mono- or polyunsaturation by 1, which has zero unsaturation. A second possible reason is the antioxidant property of the sulfur moiety in the crude and purified product mixtures. Improvements in the oxidation stability properties of oils with sulfur moieties have been reported previously and have been attributed to the antioxidation property of the sulfur group.16,33,34 Examination of Figure 3 shows slightly higher onset and peak temperatures for the crude than for the purified product mixture, which corresponds to a slightly better oxidation stability for the crude product mixture. This difference cannot be attributed to differences in sulfur content between the two products, because they are about equal (Table 3). However, it might be due to other properties related to compositional differences. As discussed before, two types of lipoyl glyceride species were generated after transesterification: hydrophilic (polar) and lipophilic (nonpolar). As shown in Tables 2 and 3, almost 50% of the species in the crude mixture are hydrophilic, whereas the corresponding value in the purified mixture is
