J Am Oil Chem Soc DOI 10.1007/s11746-013-2239-7

ORIGINAL PAPER

Synthesis and Characterization of Acetylated and Stearylyzed Soy Wax Linxing Yao • JunYi Lio • Tong Wang Darren H. Jarboe

•

Received: 30 November 2012 / Revised: 29 January 2013 / Accepted: 13 March 2013 Ó AOCS 2013

Abstract A solvent-free synthesis method was developed for incorporating acetyl and hydroxy groups and long-chain fatty alcohol in fully hydrogenated soybean oil (FHSO) to produce a wax to be used as beeswax or paraffin substitutes in packaging and coatings. FHSO was reacted with stearyl alcohol and triacetin at 140–150 °C for 2 h, and the reaction was catalyzed by 0.018 wt% sodium methoxide. The addition of alcohol increased the reaction yield and the melting point of the final wax by 9 and 25 %, respectively, compared to the wax produced from the reaction of FHSO and triacetin alone. The effects of the ratio of stearyl alcohol and triacetin to FHSO on the textural properties of the modified soy wax were examined. The reaction of FHSO:stearyl alcohol:triacetin at a molar ratio of 9:7:15 produced a wax that was 1.2 and 2.4 times harder than beeswax and FHSO, respectively, but 1.7 times softer than a commercial grade paraffin wax. This modified soy wax comprised 75 % acetylated glycerol esters including diacetylmonoacylglycerides (31 %), monoacetylmono- (12 %) and diacylglycerides (32 %), and 25 % non-acetylated molecules including wax ester (14 %), and acylglycerides (11 %). The acetylated soy wax had high cohesiveness and did not break under compression. Keywords Acetylated soy wax  Cohesiveness  Fracturability  Fully hydrogenated soybean oil  Hardness  Stearyl alcohol  Triacetin L. Yao  J. Lio  T. Wang (&) Department of Food Science and Human Nutrition, Iowa State University, 2312 Food Science Building, Ames, IA 50011, USA e-mail: [email protected] D. H. Jarboe Center for Crops Utilization Research, Iowa State University, 2312 Food Science Building, Ames, IA 50011, USA

Introduction North American wax consumption is estimated at approximately 3 billion pounds (1.36 billion kg) a year [1]. The major applications of waxes are packaging materials and candles, followed by building materials. Other uses include corrugated board, coatings, food, cosmetics and pharmaceuticals, crayons, and inks. Wax sources can be generally categorized as petroleum-based, synthetic, and natural waxes, with petroleum waxes dominating the market ([85 %). However, because of environmental concerns, depleting oil reserves, and the highly variable price of crude petroleum, petroleum waxes have gradually lost market share in some applications. Natural waxes (plant and animal derived) have the advantages of being renewable, biodegradable, and non-toxic, and are particularly important in areas where waxes are in contact with food and humans. Among natural waxes, beeswax is one of the most economically important types because of its excellent physical and textural properties. However, colony collapse disorder of honeybees [2] has markedly decreased the availability of beeswax, which has increased the cost. Other natural waxes such as carnauba and candelilla waxes have limited supplies, although they have highly desirable performance characteristics. Thus, there is a continuously increasing need for alternative and renewable solutions for replacement of these traditional materials. Wax derived from vegetable oil has been an interest of innovators because of its relatively abundant supply and low price. However, modifications of vegetable oil are needed to improve its functionality. Fully and partially hydrogenated soybean oils, referred to as soy wax, are often either too hard and brittle or greasy. Previous studies have shown that incorporating hydroxy groups, branch chains, and short-chain fatty acids (C2–C4) could significantly

123

J Am Oil Chem Soc

improve the cohesiveness of the wax [3–6]. By physically mixing acetylated monoacylglycerides and partial acylglycerides, a wax that has textural properties similar to beeswax was obtained [6]. However, a direct one-pot synthesis method for acetylated wax from vegetable oil is needed to reduce the production cost. Acetylated monoacylglycerides have been used commercially in food, cosmetics, and pharmaceuticals for many years because of their unusual crystalline properties [7, 8]. A number of patents disclosed the procedures for its preparation, which has been thoroughly reviewed [4, 9]. Interesterification of triacetin and fully hydrogenated vegetable oil is one of the most convenient methods for the synthesis of acetylated fats, but the main challenge is the poor solubility or miscibility of these two substrates, which usually requires the use of solvents, high levels of catalyst, or high shear mixing and long reaction time [4]. The objective of this study was to develop a simple and solvent-free synthesis method for incorporating acetyl, hydroxy groups, and long-chain fatty alcohol in fully hydrogenated soybean oil (FHSO), and to examine the effect of the reactant ratio on the physical and textural properties of the wax. Beeswax and commercial grade paraffin waxes as well as FHSO were used as controls for comparisons with the new soy based waxes.

Materials and Methods FHSO was provided by Stratas Foods (Memphis, TN, USA). Stearyl alcohol, triacetin, and other chemicals were purchased from Sigma-Aldrich (St. Louis, MO, USA). Beeswax was purchased from Bluecorn Naturals (Ridgway, CO, USA). Two commercial grade, fully-refined paraffin waxes used for cardboard box coating, RMP and MMP wax, were provided by Robert Mann Packaging (Salinas, CA, USA) and Michelman (Cincinnati, OH, USA), respectively. Syntheses Stearyl alcohol (20 g or 0.07 mol) was added in a roundbottom flask and heated to 100 °C under vacuum (*0.1 torr) for 15 min to remove any possible moisture, followed by the addition of the catalyst sodium methoxide (0.5 mL, 5.4 M in methanol, approximately 0.018 wt% of FHSO) and another 15 min of heating under vacuum to remove methanol. Then, FHSO (80 g or 0.09 mol) and triacetin (33 g or 0.15 mol) were quickly added to the alcohol. The reaction mixture was heated for 20 min to melting under vacuum. Then, the temperature was increased to 140–150 °C under N2. After the 2 h reaction with continuous stirring at 300 rpm, the wax was cooled to

123

about 110 °C, and then 2 mL of hydrochloric acid (HCl) was added drop-wise, followed by 1,000 ppm of citric acid (dissolved in 1 mL water) as described in a previous study [4]. After 30 min stirring, the wax was poured into 700 mL hot water (70 °C) to remove any unreacted triacetin, diacetin, monoacetin, possible trace glycerol, soaps, salts, and excess acid. The liquid wax in the hot water was vigorously stirred using a spatula and then left at 0 °C until the wax solidified. The solid wax was then washed 7 times using fresh hot water. The residual moisture in the wax was removed by a rotary evaporator at 80 °C for 30 min. To examine the effects of the addition of triacetin and stearyl alcohol on the chemical and physical properties of the final product, triacetin was added at 0, 0.025, 0.05, 0.075, 0.1, 0.125, 0.15, and 0.2 mol to fixed amounts of FHSO (0.09 mol) and stearyl alcohol (0.07 mol). Then, to a fixed amount of FHSO (0.09 mol) and triacetin (0.15 mol), stearyl alcohol was added at 0, 0.04, 0.07, 0.11, and 0.15 mol. Three synthesis replicates were done for each treatment. Chemical Analyses The incorporation of acetyl group in the wax was determined using 1H NMR spectroscopy. The 1H NMR spectra were obtained using a Varian MR-400 spectrometer (400 MHz; Santa Clara, CA, USA) with OneNMR pulsefield-gradient probe at 25 °C. The MNova software (Mestrelab Research, Escondido, CA, USA) was used for data processing. All samples were dissolved in chloroformd (CDCl3). The signal of C3 protons on the long fatty acid moiety and on the alcohol moiety shifts at 1.61 ppm. The methyl protons on the acetyl group have signals at 2.07–2.16 ppm. The incorporation of acetyl group was expressed as the molar ratio of acetyl group to the total alkyl group (i.e., long fatty acid and alcohol chains and acetyl group), which was calculated as the area integration ratio of the two signals: % Incorporation of acetyl group = 0.33 9 Integration2.07-2.16/(0.33 9 Integration2.07-2.16 ? 0.5 9 Integration1.61), where the subscript numbers are the chemical shifts of peaks being integrated. Two NMR samples were analyzed per batch of wax. To determine the wax molecular composition, a wax sample (about 30 mg) from the treatment of 0.15 mol triacetin and 0.07 mol alcohol was dissolved in CHCl3 and applied on a preparative TLC plate (Silica gel G, 500 lm thickness; Analtech, Newark, DE, USA). The plate was developed in a solvent mixture of hexane/diethyl ether/ acetic acid (70:30:1, by vol). The separation was visualized under UV light after spraying with 0.1 % (w/v) 20 , 70 dichlorofluorescein in methanol. The silica bands were scraped off and extracted twice with 10 mL diethyl ether. The combined ether extract was evaporated under a stream

J Am Oil Chem Soc

testing (5 replicates) were made using the bottom of a pipette tip rack as a mold for the desired cubic shape. The cooled and solidified samples were stored at ambient temperature for 24 h before analysis. For the bend test (with 3 replicates of subsampling), 6 g of wax was melted in an aluminum dish, and then a flat round disk (55 mm diameter 9 3 mm thickness) was formed upon cooling to ambient temperature.

of nitrogen to dryness and then the purified components were dissolved in CDCl3 for structural analysis by 1H NMR. For quantification after the NMR identification of the individual bands, another three replicates were prepared using the method described above and the resultant ether extracts after drying were added with 2 mL of 2 % H2SO4 in methanol in the presence of an internal standard nonadecanoic acid. After a 20-h incubation at 60 °C, the resultant fatty acid methyl esters (FAME) were extracted by hexane (3 mL) and washed with water (10 mL). The hexane solution (1 lL) was injected into a GC equipped with a SPB 2330 column (Supelco, St. Louis, MO, USA). The temperature program started at 150 °C for 1 min and then increased to 220 °C at a rate of 10 °C/min. The column flow rate of helium was 1 mL/min and the split ratio was 25. The area percentage of GC peaks (C16:0, C18:0, and internal standard) was used to calculate the mass of C16:0 and C18:0 in each TLC fraction. Then, total moles of C16:0 and C18:0 were determined. The mass of the main component in each TLC band was calculated by multiplying their moles by the molecular weights listed in Table 1. Overall, four TLC fractionations (from one wax) were done, one used for NMR structural analysis, and three used for replicates of GC analysis.

Textural Analysis A TA.XT2i Texture Analyzer (Stable Micro Systems, Godalming, UK) was used to measure the hardness, cohesiveness/brittleness, and fracturability of the waxes. Hardness was measured by a penetration test with a TA212 cylinder probe (3 mm diameter). The test speed was 0.2 mm/s, and the probe traveled 5 mm into the wax. The peak force (g) in the penetration test was defined as the hardness. A compression test was conducted with a TX plate probe (76 mm diameter). The cubic wax samples (8 9 8 9 8 mm) were compressed for a distance of 6 mm to give 75 % deformation. Visual and qualitative examinations of the deformation or disintegration were recorded. A custom-built three-point accessory was used in the bend test. The wax disk was placed on two vertical support bars 14 mm apart. A third bar attached to the crosshead of the instrument was driven perpendicular into the sample at a speed of 0.5 mm/s, with 10 mm of travel distance. The area under the force curve before the wax was broken was recorded as the wax resistance to fracture.

Sample Preparation for Textural Analysis The wax was heated to melting and then about 16 g of hot wax was poured into a tin container (3.81 cm diameter, 2.54 cm height) for the penetration test with three replicates of subsampling or testing. The samples for compression

Table 1 The chemical composition of the wax determined by TLC, 1H NMR and GC Band #

1

2

3

4

5

6

7

8

9

Rf wt% of total wax

0.68 13.6 ± 1.0

0.57 3.7 ± 0.8

0.45 10.3 ± 0.4

0.39 21.8 ± 0.8

0.24 31.2 ± 0.8

0.18 4.0 ± 0.1

0.13 2.1 ± 0.1

0.07 11.9 ± 0.6

0 1.4 ± 0.1

Molecular weighta

532

872

658

658

438

616

616

396

354

RCOOSb

The assayed wax was from the treatment of 0.09 mol of FHSO, 0.07 mol of stearyl alcohol, and 0.15 mol of triacetin. The stereoisomers of sn-1 and sn-3 were not differentiated a An averaged molecular weight of the long chain fatty acid, 280, was used to calculate the molecular weight of the main component in each TLC fraction b

S stearyl alcohol, R long chain alkyl group (of stearic and palmitic acid), Ac acetyl group

123

J Am Oil Chem Soc

Statistical Analysis Three batches of waxes were synthesized for each treatment. Each batch of wax was used to provide 3 replicates for the penetration test, 5 replicates for the compression test, 3 replicates for the bend test, 2 replicates for melting point determination, and 2 replicates for 1H NMR analysis. The treatment effects were examined at the 5 % significance level by fitting the data to a linear mixed model using the restricted maximum likelihood (REML) method and Bonferroni correction and analyzed by Statistical Analysis System (SAS) 9.1 (SAS Institute, Cary, NC, USA). The means and standard deviations were determined and presented.

Results and Discussion Incorporation of Acetyl Group and Reaction Yield as Affected by the Addition of Triacetin and Stearyl Alcohol The incorporation of acetyl group measures the amount of acetyl group relative to total long and short alkyl chains in the wax. As shown in Fig. 1a, the amount of acetyl group in the synthesized wax increased with the addition of triacetin, except for the treatment with the highest amount of triacetin. Because of the solubility problem, two-component reaction of FHSO and triacetin (without solvent) usually cannot achieve a high yield under mild reaction conditions [4]. Long chain alcohol having an intermediate polarity can help solubilize triacetin in FHSO, and, thus, facilitate the reaction. The amount of triacetin that is

123

Incorporation of acetyl group (%)

The melting and crystallization properties of the waxes were analyzed using a differential scanning calorimeter (DSC-7; Perkin-Elmer, Norwalk, CT, USA) equipped with an Intracooling II system. Solid wax (3–5 mg samples) was weighed in an aluminum pan (Perkin-Elmer) and sealed. Indium was used as the reference standard. The temperature program started with a 1-min hold at 25 °C, followed by 40 °C/min rapid heating to 85 °C and a 5-min hold at 85 °C. Then the sample was cooled to -10 °C at 10 °C/ min. After a 15-min hold at -10 °C, the sample was heated to 85 °C at 5 °C/min. The completion of melting or melting point was collected during the last heating step and calculated using Pyris software (Perkin-Elmer) as the intersection of tangent lines drawn from the high temperature side of the melting peak of the highest temperature and the baseline. Two replicates from each wax sample were measured.

(a) 50 40 30 20 10 0 0

0.05

0.1

0.15

0.2

0.25

Addition of Triacetin (mol)

(b) Incorporation of acetyl group (%)

Thermal Analysis

50 40 30 20 10 0 0

0.05

0.1

0.15

0.2

Addition of Alcohol (mol) Fig. 1 Incorporation of acetyl group in the wax (mol %) determined by 1H NMR. a Effect of the addition of triacetin to 0.09 mol of FHSO and 0.07 mol of stearyl alcohol. b Effect of the addition of stearyl alcohol to 0.09 mol of FHSO and 0.15 mol of triacetin. Means with the same letters are not significantly different at a = 0.05 (n = 3). Error bars standard deviations

soluble in 80 g of FHSO can be increased from 5 to 30 g in the presence of 20 g stearyl alcohol. Except for the treatment containing 0.2 mol triacetin (the highest level), other treatments all had one reaction phase (transparent liquid before adding catalyst). On the other hand, with more alcohol added, the proportion of triacetin as a substrate in the starting reaction mixture became less, resulting in a decrease in the amount of acetyl group in the wax (Fig. 1b). The highest percentage of acetyl group that was incorporated in the wax was about 40–42 % based on a molar ratio to total alkyl groups. To a fixed amount of FHSO and stearyl alcohol, the reaction yield decreased from 94.5 to 81.3 % with more triacetin added (Table 2). To a fixed amount of FHSO and triacetin, the yield increased with the addition of alcohol, from 80.5 to 88.0 %. Due to the large error of one treatment, the mean comparison of treatments with alcohol as the variable

J Am Oil Chem Soc Table 2 Reaction yield and melting point as affected by substrate ratio Stearyl alcohol (mol)

Table 3 Physical properties of FHSO, commercial paraffin waxes (MMP and RMP), and beeswax

Triacetin (mol)

Reaction yield (%)

Melting point (°C) FHSO

1,664 ± 282c

2,555 ± 286c

65.48 ± 0.01b

0.07

0

94.5 ± 3.1 a

55.05 ± 2.41 a

MMP

6,712 ± 347a

8,192 ± 325c

62.77 ± 0.04c

0.07 0.07

0.025 0.050

93.4 ± 3.9 ab 93.5 ± 2.6 a

50.99 ± 0.25 b 50.66 ± 2.06 bc

RMP Beeswax

6,890 ± 279a 3,362 ± 66b

22,895 ± 2,751b 38,115 ± 2,864a

61.54 ± 0.08d 66.24 ± 0.02a

Mean ± SD with the same letters are not significantly different at a = 0.05 (n = 3)

Effect of triacetin

0.07

0.075

91.9 ± 1.7 abc

48.99 ± 0.28 bc

0.07

0.100

84.4 ± 3.0 bcd

47.56 ± 1.20 c

0.07

0.125

84.4 ± 4.2 bcd

48.15 ± 0.38 c

0.07

0.150

84.3 ± 2.0 cd

47.36 ± 0.54 c

0.07

0.200

81.3 ± 2.2 d

47.99 ± 0.80 c

80.5 ± 1.1 a

40.31 ± 1.15 b

Hardness (g)

0.04

0.15

82.1 ± 5.3 a

41.41 ± 2.97 b

0.07

0.15

84.3 ± 2.0 a

47.36 ± 0.54 a

0.11

0.15

87.6 ± 0.8 a

49.33 ± 0.59 a

0.15

0.15

88.0 ± 1.3 a

50.69 ± 0.18 a

10000

Force (g)

0.15

Melting point (°C)

(a) 12000

Effect of stearyl alcohol 0

Resistance to fracture (g mm)

Mean ± SD with the same letters are not significantly different at a = 0.05 (n = 3)

8000 6000 4000 2000 0

0

Textural Properties as Affected by the Addition of Triacetin and Stearyl Alcohol Hardness is an important property of wax used in box and paper coating applications. A harder wax is usually preferred. The hardness of FHSO is about 1,664 g, much lower than beeswax (3,362 g) or commercial paraffin waxes (6,712-6,890 g) (Table 3). Without stearyl alcohol, the addition of triacetin increased the hardness of the wax to 4,902 g (Fig. 2). Without triacetin, stearyl alcohol made the wax even harder (5,943 g). However, waxes obtained from these two treatments had unacceptable appearances as shown in Fig. 3 (left) that some components have separated and ‘‘bloomed’’ on the surface. It was also found that such two-component treatments had large batch-to-batch variations in terms of textural properties of the final products. These defects might be caused by the immiscibility of reactants and the difference in hydrophobicity of the components in the product. Adding a small amount of triacetin (0.025 or 0.05 mol) to FHSO and stearyl alcohol increased the hardness by twofold, which made the wax even harder than the commercial paraffin waxes. When more triacetin was added ([0.05 mol), hardness of the waxes decreased. There was a general trend of decreasing hardness with an increasing amount of alcohol added, but

0.05

0.1

0.15

0.2

0.25

Addition of Triacetin (mol)

(b) 12000 10000

Force (g)

showed that the yields were not statistically different. Further study is needed to understand the kinetics and equilibrium of such a three-component reaction.

8000 6000 4000 2000 0

0

0.05

0.1

0.15

0.2

Addition of Alcohol (mol) Fig. 2 Hardness determined by a penetration test. a Effect of the addition of triacetin to 0.09 mol of FHSO and 0.07 mol of stearyl alcohol. b Effect of the addition of stearyl alcohol to 0.09 mol of FHSO and 0.15 mol of triacetin. Means with the same letters are not significantly different at a = 0.05 (n = 3). Error bars standard deviations

the degree of the reduction was small. Except for the treatments with the highest level of triacetin or stearyl alcohol, all other waxes were harder than beeswax. As shown in Fig. 3 (right), the addition of triacetin had a great effect on the cohesiveness of the wax. When the amount of triacetin added was less than 0.075 mol, the waxes were brittle and broke into pieces under compression. Adding 0.075–0.125 mol of triacetin reduced the

123

J Am Oil Chem Soc

Area (g·mm)

(a) 40000 30000

20000

10000

0 0

0.05

0.1

0.15

0.2

0.25

Addition of Triacetin (mol)

Area (g·mm)

(b) 40000 30000

20000

10000

0 0

Fig. 3 Examples of ‘‘bloomed’’ waxes made from the treatment of FHSO and 0.15 mol of triacetin (upper left) and the treatment of FHSO and 0.07 mol of stearyl alcohol (lower left), and the deformation of waxes under compression (right). The numbers on the right picture are the moles of triacetin added to FHSO (0.09 mol) and stearyl alcohol (0.07 mol)

brittleness so that the waxes formed a flat disk from a cubic shape and retained their integrity. Nonetheless, small cracks were found on the edges of these samples. The treatment of 0.15 mol of triacetin gave the best appearance after compression, that is, the wax deformed without any cracking, the same as observed for beeswax. Increasing the amount of stearyl alcohol did not affect the appearance of the wax that already had 0.15 mol triacetin added, and their pictures are not presented. The three-point bend test has been used in the textural analysis of foodstuffs such as biscuits, potato chips, and cheese for determining their fracturability [10]. We recorded the maximum break force, distance that the probe travelled, and the area under the force–distance curve, and found that the area was the most sensitive parameter to differentiate the samples. The area value in Fig. 4 is the work to fracture by bending, a higher value meaning lower fracturability and higher resistance to fracture. With more triacetin added, the ability of the wax to resist fracture became greater (Fig. 4). The highest level of triacetin addition had large variation from batch to batch in the bend test, although the subsampling replicates agreed well with

123

0.05

0.1

0.15

0.2

Addition of Alcohol (mol) Fig. 4 The resistance to fracture determined by a bend test. a Effect of the addition of triacetin to 0.09 mol of FHSO and 0.07 mol of stearyl alcohol. b Effect of the addition of stearyl alcohol to 0.09 mol of FHSO and 0.15 mol of triacetin. Means with the same letters are not significantly different at a = 0.05 (n = 3). Error bars standard deviations. The treatment with open diamond in (a) having large variation was discarded in data analysis

each other. This is likely caused by the non-uniform substrate mixing and reaction as discussed earlier. The effect of stearyl alcohol on fracturability is different from that of triacetin. There seemed to have been an optimal amount of alcohol (0.075 and 0.11 mol) that produced waxes with greater resistance to fracture, about 18,000–19,000 g mm. This range value was lower than beeswax and RMP wax ([20,000), but higher than MMP and FHSO (\10,000) (Table 3). The optimal amount of alcohol for producing the wax with low fracturability may be a function of a hydroxy group because it is known to increase the cohesiveness of waxes [5]. Melting Point as Affected by the Addition of Triacetin and Stearyl Alcohol The melting points (completion of melting) of the acetylated soy waxes ranged from 40.3 to 55.1 °C (Table 2). Beeswax, commercial grade paraffin waxes, and FHSO all melted above 60 °C (Table 3). The melting point of

J Am Oil Chem Soc

modified waxes decreased with the addition of triacetin and increased with the addition of stearyl alcohol. A high melting point is usually desirable for coating and packaging waxes. To further improve the melting point, commercially used additives such as various grades of polyethylene products may be added to formulate final products. The DSC melting profile in Fig. 5 shows that two fractions, high-melting and low-melting components, were present in the waxes and their relative proportion changed with the reactant ratio. Without triacetin, the wax had a single smooth peak at about 53 °C, which can be attributed to the wax esters and partial acylglycerides with long and saturated acyl chains. With more triacetin added, this peak shifted to lower temperatures and became smaller, while a peak at about 37 °C appeared and became dominant (Fig. 5a). As mentioned earlier, the reaction of triacetin and FHSO was inconsistent without stearyl alcohol. The top curve in Fig. 5b is an example of an incomplete reaction with unreacted triacylglycerides (TAG) and perhaps some partial glycerides giving a high melting peak at *50 °C. We also observed another batch of wax (from triacetin and FHSO alone) that was much more resistant to fracture than other replicates and gave one single smooth DSC peak, indicating reaction variability. With a small amount of alcohol added, the TAGs were fully interesterified with triacetin and alcohol, and the wax had one single peak in the low temperature region (the second curve from

Chemical Characterization of the Wax by TLC, GC, and 1H NMR The wax applied onto a silica gel plate was developed to 9 bands. Their chemical structures, Rf values, and percent relative to total wax are summarized in Table 1. The characterization results discussed in this section were from one wax sample (obtained from the reaction of 0.09 mol FHSO, 0.07 mol stearyl alcohol, and 0.15 mol triacetin), so the composition of other waxes might vary with their treatment conditions. Band 1 with the highest Rf value was the wax ester from palmitic or stearic acid and stearyl alcohol. 1H NMR analysis obtained signals at d 4.06 (2H, t), 2.30 (2H, t), 1.65 (2H, m), and 0.88 (3H, t) ppm. Signals for the protons of methyl esters and acetyl groups were not observed in this fraction. It is possible that stearyl acetate was present in other wax samples, maybe in low concentration. Band 2 contained TAG with palmitic and stearic acid. Bands 3 and 4 both contained monoacetyldiacylglycerides, with a type ABA in band 3 and type AAB in band

(b)

Endo up (mW)

(a)

Endo up (mW)

Fig. 5 The DSC melting profile of the soy waxes. a Effect of the addition of triacetin to 0.09 mol of FHSO and 0.07 mol of stearyl alcohol. b Effect of the addition of stearyl alcohol to 0.09 mol of FHSO and 0.15 mol of triacetin. The numbers are the mol of triacetin (a) and stearyl alcohol (b) added

the top of Fig. 5b). With more long chain alcohol added, the high-melting fraction appeared and its peak became larger. The high-melting fraction was likely to be the mixture of wax ester, TAG, diacylglycerides (DAG), and monoacylglycerides (MAG), and the low-melting fraction was the acetylated glycerol esters.

Temperature (ºC)

Temperature (ºC)

123

J Am Oil Chem Soc

4, where the letters representing fatty acids denote the position of the fatty acyl groups on the sn-1, 2, and 3 positions of the glycerol backbone. Both bands showed multiplets at 5.26 ppm for the proton on the sn-2 carbon and a singlet at 2.07 ppm for the methyl protons on the acetyl group. Band 3 (ABA) had two double doublets at 4.13–4.32 ppm that arose from shifts of the identical protons on the sn-1 and sn-3 carbons. Band 4 also had two sets of doublets in the same region, with one set in the up-field region as double doublets, but the other set in the downfield region had multiple doublets and was not well resolved. The other spectral difference of the two bands was found in the region of C2 protons. Band 3 had a triplet at 2.32 ppm, whereas band 4 had multiple triplets at 2.29–2.35 ppm. The spectral characteristics of C2 protons can be used to distinguish between ABA and AAB TAGs having short chain acyl moieties with chain length up to six carbon atoms [11]. A singlet peak at 2.07 ppm was assigned to the protons on an acetyl group. The integral ratios of C2 protons to acetyl protons were 4.5:3 and 4.2:3 for bands 3 and 4, respectively. Band 5 contained diacetylmonoacylglycerides. It had an almost identical spectrum as band 4 except for the slight difference in the splitting pattern in the region of 4.13–4.32 ppm and an integral ratio of 1.2:3 for C2 protons to acetyl protons. It may be a mixture of AAB and ABA. Given the similar spectral properties to triacetin [11], a doublet at 2.07 ppm and a singlet at 2.09 ppm in a ratio of 2:1 were assigned to the methyl protons on the acetyl group. Band 6 was 1,3-DAG, and its spectrum agreed well with the literature [12, 13]. The two double doublets in the region of 4.14–4.23 ppm were assigned to the a proton (4H) on the glycerol backbone. The multiplets (1H) at 4.06–4.10 ppm were attributed to the b proton (12). The OH proton at the sn-2 position gave a set of doublets at 2.44 ppm. The integration ratio of a, b, and OH protons was 4:1.2:1. Band 6 also contained less than 5 % stearyl alcohol in the assayed wax sample as detected by GC. Band 7 was 1,2-DAG. The characteristic shift at 5.09 ppm was attributed to the b proton [14]. The a proton at the sn-1 carbon remained at 4.14–4.23 ppm, but the proton at the sn-3 position shifted to 3.74 ppm. The integration ratio of protons at sn-1, 2, and 3 was 2.2:1:1.8. The shift of C2 protons at 2.3 ppm was a quartet. Band 8 was monoacetylmonoacylglycerides. The spectrum showed two singlet shifts at 2.08 and 2.11 ppm for the methyl protons on the acetyl group in a ratio of 1.1:5. The integration ratio of C2 protons to methyl protons on the acetyl group was 2:3. The shift at 5.07 ppm was assigned to the b proton of the 1,2-isomer and its ratio to the C2 protons was 1:4.2, which suggested that band 8 contained an isomeric mixture of monoacetylmonoacylglycerides of

123

both ABOH and AOHB types. The protons on the sn-3 carbon of the 1,2-isomer shifted to 3.75 ppm. Other protons on the glycerol backbone gave overlapped multiplets at the region of 4.07–4.35 ppm. The long acyl chain and acetyl group at sn-1 or 3 in AOHB and at sn-1 or 2 in ABOH may result in different spectral properties, and may be distinguished by 13C NMR analysis [15]. The current report does not include such an analysis. Band 9 contained a mixture of 1- and 2-MAG in a 9.1:1 ratio determined by 1H NMR. This ratio is similar to the 9:1 ratio obtained in previous studies when acyl migration of MAG reached equilibrium [13, 16]. The spectral peak shifts and assignments of the glycerol backbone protons of 1- and 2-MAG correlated well to those previously reported [12, 13]. The two protons on the sn-1 carbon of 1-MAG gave double doublet signals at the region from 4.13 to 4.23 ppm. The protons at sn-2 and sn-3 gave multiplet signals at 3.93, 3.70, and 3.62 ppm. The integration of the peaks was attributed to protons attached to sn-1, 2, and 3 carbon of 1-MAG was 2.2:1:2.1. Two sets of multiplets at 4.93 and 3.84 ppm were assigned as the protons on the sn-2 and sn-1/3 carbons of 2-MAG in a 1:3.8 ratio. The integration of protons at the sn-2 carbon of the two isomers was used to determine their ratio in the wax. TLC-GC analysis showed that molecules containing acetyl groups were 75.2 % (by weight) of the wax synthesized with 0.09 mol FHSO, 0.07 mol stearyl alcohol, and 0.15 mol triacetin, with the most abundant species being monoacetyldiacylglycerides (32.1 %, bands 3 and 4) followed by the diacetylmonoacylglycerides (31.2 %, band 5) and then monoacetylmonoacylglycerides (11.9 %, band 8). The AAB type (band 4) of monoacetyldiacylglycerides was favored in this non-specific interesterification, being twice as much as the ABA type. Among the molecules not containing acetyl groups, wax ester (band 1), long acyl chain TAG (band 2), DAG (bands 6 and 7), and MAG (band 9) were 13.6, 3.7, 6.1, and 1.4 % of the total, respectively. The ratio of 1,3-DAG (band 6) to 1,2-DAG (band 7) was 1.9:1, consistent with the 2:1 ratio at the equilibrium state of two isomers in the literature [14]. The novelty of this successfully demonstrated method for the synthesis of acetylated and stearylyzed soy wax is the use of stearyl alcohol and its beneficial effect for both the reaction and product quality. The long chain alcohol not only facilitated the reaction between FHSO and triacetinbut also contributed to the textural properties of the modified wax. Stearyl alcohol may be replaced by other alcohols or a mixture of alcohols with different chain lengths, which will probably result in waxes with different textural properties for a variety of applications. Of the three reactants, FHSO and triacetin are generally regarded as safe, and stearyl alcohol is permitted as a food additive for direct addition to food for human consumption [17].

J Am Oil Chem Soc

Therefore, the modified wax from this solvent-free synthesis has great potential for edible applications such as fruit, vegetable, and cheese coatings, as well as packaging, box, and paper coatings that are in direct contact with food. Our preliminary coating comparison with commercial paraffin products has shown similar performance. Various in-depth coating performance evaluations are a topic for future study. In summary, this wax synthesis introduces a new and simple method for producing acetylated and stearylyzed soy wax that has significantly improved textural properties. Long chain fatty alcohol increased the yield of the FHSO and triacetin reaction. The incorporation of acetyl and hydroxy groups is important for reducing the brittleness of FHSO and the incorporation of the long chain fatty alcohol group is important for increasing melting point. This solvent-free wax synthesis procedure should be readily scalable because of the inexpensive raw materials and low energy consumption attributed to the mild reaction conditions. Thus, it has a great potential to be adopted commercially as a green processing technology to produce high performance, eco-friendly, and non-toxic base wax materials for various applications. Acknowledgments We thank United Soybean Board for financial support, Stratas Foods for supplying FHSO material, and Charlwit Kulchaiyawat for assistance in collecting NMR data.

References 1. American Fuel and Petrochemical Manufacturers (accessed Nov 2012) Wax facts. http://www.afpm.org/wax-facts/ 2. Leal WS, Bromenshenk JJ, Henderson CB, Wick CH, Stanford MF, Zulich AW, Jabbour RE, Deshpande SV, McCubbin PE, Seccomb RA, Welch PM, Williams T, Firth DR, Skowronski E, Lehmann MM, Bilimoria SL, Gress J, Wanner KW, Cramer RA Jr (2010) Iridovirus and microsporidian linked to honey bee colony decline. PLoS One 5:e13181

3. Feuge RO, Vicknair EJ, Lovegren NV (1952) Modification of vegetable oils. XII. Plasticity of some aceto derivatives of monostearin. J Am Oil Chem Soc 29:11–14 4. Pelloso TA, Roden AD, Boldt GL (1995) Synthesis of acetoglyceride fats. US Patent 5,434,278 5. Wang L, Wang T (2007) Chemical modification of partially hydrogenated vegetable oil to improve its functional properties for candles. J Am Oil Chem Soc 84:1149–1159 6. Yao L, Wang T (2012) Textural and physical properties of biorenewable ‘‘waxes’’ containing partial acylglycerides. J Am Oil Chem Soc 89:155–166 7. Alfin-Slater RB, Coleman RD, Feuge RO, Altschul AM (1958) The present status of acetoglycerides. J Am Oil Chem Soc 35:122–127 8. Young FVK, Poot C, Biernoth E, Krog N, Davidson NGJ, Gunstone FD (1994) Processing of fats and oils. In: Gunstone FD, Harwood JL, Padley FB (eds) The lipid handbook. Chapman & Hall, London 9. Dijkstra AJ, Brocke PC, Woldhuis J (2006) Process for preparing an acetoglyceride composition comprising high amounts of monoacetyl monoacyl glycerides. WIPO patent application WO/ 2006/091096 and European patent application EP1838656 10. Stable Micro systems (accessed Nov 2012) Bakery product testing attachments. http://www.stablemicrosystems.com/bakeryapps.htm 11. Lie Ken Jie MSF, Lam CC (1995) 1H-nuclear magnetic resonance spectroscopic studies of saturated, acetylenic and ethylenic triacylglycerols. Chem Phys Lipids 77:155–171 12. The Lipid Library (accessed Nov 2012) 1H NMR spectroscopy of fatty acids and theirderivatives: glycerol esters. http:// lipidlibrary.aocs.org/nmr/1NMRglyc/index.htm 13. Compton DL, Vermillion KE, Laszlo JA (2007) Acyl migration kinetics of 2-monoacylglycerols from soybean oil via 1H NMR. J Am Oil Chem Soc 84:343–348 14. Laszlo JA, Compton DL, Vermillion KE (2008) Acyl migration kinetics of vegetable oil 1, 2-diacylglycerols. J Am Oil Chem Soc 85:307–312 15. Gunstone FD, Shukla VKS (1995) NMR of lipids. Annu Rep NMR Spectrosc 31:219–237 16. Boswinkel G, Derksen JTP, van’t Riet K, Cuperus FP (1996) Kinetics of acyl migration in monoglycerides and dependence on acyl chain length. J Am Oil Chem Soc 73:707–711 17. U.S. Food and Drug Administration (accessed Nov 2012) Code of Federal Regulations Title 21. http://www.fda.gov/Food/Food IngredientsPackaging/FoodAdditives/FoodAdditiveListings/ucm 091048.htm#ftnS

123