Solubility Testing of Sucrose Esters of Fatty Acids in International Food

284 Biol. Pharm. Bull. 40, 284–289 (2017) Vol. 40, No. 3 Regular Article

Solubility Testing of Sucrose Esters of Fatty Acids in International Food Additive Speciﬁcations Yukino Nagai,a Satoko Kawano,*,a Kenichiro Motoda,b Masaaki Tomida,a,† Chiye Tatebe, c Kyoko Sato, c and Hiroshi Akiyamac a Food Ingredients Department, Performance Products Division, Mitsubishi Chemical Corporation; 1–1–1 Marunouchi, Chiyoda-ku, Tokyo 100–8251, Japan: b Research and Development Center, Mitsubishi-Kagaku Foods Corporation; 1000 Kamoshida-cho, Aoba-ku, Yokohama 227–8502, Japan: and c National Institute of Health Sciences; 1–18–1 Kamiyoga, Setagaya-ku, Tokyo 158–8501, Japan. Received September 15, 2016; accepted December 13, 2016; advance publication released online December 20, 2016

We investigated the solubility of 10 samples of sucrose esters of fatty acids (SEFA) products that are commercially available worldwide as food additives (emulsifiers). Although one sample dissolved transparently in both water and ethanol, other samples produced white turbidity and/or precipitates and did not meet the solubility criterion established by the Joint Food and Agriculture Organization of the United Na- tions (FAO)/WHO Expert Committee on Food Additives (JECFA). When the sample solutions were heated, the solubility in both water and ethanol increased. All of the samples dissolved transparently in ethanol, and dispersed and became white without producing precipitates in water. The present study suggests that the current solubility criterion of the JECFA SEFA specifications needs to be revised. Key words sucrose fatty acid ester; solubility; specification; emulsifier; food additive

Sucrose esters of fatty acids, in which hydroxyl groups are esters of fatty acids” (Table 1). esterified with fatty acids, are used worldwide as emulsifiers In this study, we use the term SEFA to indicate that the and stabilizers in food, cosmetic, and pharmaceutical prod- product is hydrophilic and composed of sucrose esters of ucts.1–4) For food additives, fatty acids derived from edible fatty acids for which the total mono- to triester content is fats and oils (lauric acid, palmitic acid, stearic acid, oleic acid, not less than 80%, as defined by the JECFA, EU, and United etc.) are used. Stearic acid-type products are preferentially States. SEFA has been approved in a number of countries and used because of their flavor and functional characteristics. are used in oil-in-water emulsion-type foods (e.g., whipped As a sucrose molecule has 8 hydroxyl groups, it can produce cream), wheat products (e.g., bread, biscuits, and cakes), and esters ranging from mono- to octaesters. Commercial products confectionaries (e.g., chewing gums and candies). produced using the methods described in the Joint Food and The JECFA first evaluated the safety of SEFA in 1969. Agriculture Organization of the United Nations (FAO)/WHO Subsequently, safety evaluations were carried out based on Expert Committee on Food Additives (JECFA)—“Sucrose es- newly available animal and human studies in the 44th (1995) ters of fatty acids” specifications5) are thought to be mixtures and 49th (1997) meetings.6,7) Solubility testing is required for of these sucrose esters. It is known that products in which the the identification of SEFA according to the JECFA specifica- monoester content is higher or the chain length of the fatty tions, and this requirement has not been changed for nearly 40 acids esterified to sucrose is shorter exhibit more hydrophilic years. Japan, China, South Korea, and Taiwan do not have a properties. A representative structure of sucrose monostearate solubility criterion in their specifications; however, solubility is shown in Fig. 1. testing in the JECFA specifications is now commonly required Sucrose esters of fatty acids is approved for use as a food when SEFA products are used in foods sold worldwide. additive in a number of countries, and the specifications for its In this study, we investigated the solubility of samples of use are established in those countries to ensure the quality and commercially available SEFA products based on the JECFA stability of the products. In particular, the specifications of the methods. Most SEFA product samples did not dissolve in ei- JECFA are considered international food additive specifica- ther water or ethanol and thus did not meet the JECFA SEFA tions. According to the specifications of the JECFA, European solubility criterion. In particular, we found that SEFA product Union (EU), and United States, “sucrose esters of fatty acids” must contain a total content of mono-, di-, and triesters of not less than 80%. A substance that has a lower mono- to triester content (i.e., that has a much higher esterified composition [tetra- or higher esters]) is designated a “sucrose oligoesters.” In Asian countries, including Japan, China, South Korea, and Taiwan, there is no restriction on ester composition, and all products are approved under the single designation “sucrose

† Present address: Plant Factory Business Development Department, Per- formance Products Division, Mitsubishi Chemical Corporation; 1–1–1 Marunouchi, Chiyoda-ku, Tokyo 100–8251, Japan. Fig. 1. Representative Structure of Sucrose Monostearate

Table 1. Classiﬁcation of Sucrose Esters of Fatty Acids

Hydrophilic Lipophilic

Japan, China, Korea, Taiwan Sucrose esters of fatty acids

JECFA Sucrose esters of fatty acids (INS473) Sucrose oligoesters type II Sucrose oligoesters type I United States Sucrose esters of fatty acids Sucrose oligoesters EU Sucrose esters of fatty acids (E473) — samples that were test materials in the safety studies evalu- Table 2. Descriptive Terms for Solubility in the JECFA Specifications ated by the JECFA did not meet the solubility criterion of the Parts of solvent required JECFA SEFA specifications. We discuss some of the factors Descriptive term for 1 part of solute that may have influenced these results. Very soluble Less than 1 MATERIALS AND METHODS Freely soluble From 1 to less than 10 Soluble From 10 to less than 30 Samples and Reagents Ten commercially available SEFA Sparingly soluble From 30 to less than 100 products were provided by Japanese and European companies. Slightly soluble From 100 to less than 1000 Six samples of RYOTO SUGAR ESTERS (L-1695, L-595, Very slightly soluble From 1000 to less than 10000 P-1670, S-1670, S-1170, and S-570) were obtained from Mitsu- Practically insoluble or insoluble More than 10000 bishi-Kagaku Foods Corporation (Tokyo, Japan). They were manufactured by Mitsubishi Chemical Corporation (Tokyo, GC (Agilent Technologies Japan, Ltd., Tokyo, Japan) equipped Japan). Two samples of DK ESTER (F-110 and F-50) were with an Agilent 7693 autosampler (Agilent Technologies obtained from DKS Co., Ltd. (Kyoto, Japan). Two samples of Japan, Ltd.) and a flame ionization detector (FID). An Agilent DUB SE (11S and 5S) were obtained from Stearinerie Dubois J&W DB-WAX column (30 m×0.53 mm i.d., 1 µm) (Agilent SA (Boulogne-Billancourt, France). S-1170 and S-570 were Technologies Japan, Ltd.) was used for the separations. SEFA test materials in the safety studies evaluated by the JECFA. samples were silylated using a Fatty Acid Methylation Kit Except for the solubility criterion, all samples met the JECFA (100 tests) (Nacalai Tesque, Inc., Kyoto, Japan). One micro- SEFA specifications. liter of sample was injected into a splitless liner using an inlet The water used was deionized (≤1 µS/cm) and obtained temperature of 240°C and a 4 mL/min nitrogen gas flow rate. from a water purification system (Classpure C-10P, Mitsubi- The FID temperature was 240°C. The percentage of fatty acid shi Rayon Aqua Solutions Co., Ltd., Tokyo, Japan). Ethanol composition in each sample was calculated by the following (99.5% (w/w)) was purchased from Wako Pure Chemical In- formula: dustries, Ltd. (Osaka, Japan). % fatty acid= 100 AT / Properties of Individual Samples 1) Analysis of Monoester Content where A is the peak area of the fatty acid, and T is the sum of The monoester contents of the 10 samples were analyzed all peak areas eluting within 60 min of initiating elution. by gel permeation chromatography (GPC) according to the Solubility Testing Solubility was investigated according JECFA test methods5) using an HLC-8320GPC (Tosoh Cor- to the JECFA test methods.9) The solubility criterion of the poration, Tokyo, Japan) equipped with a refractive index JECFA describes SEFA as “sparingly soluble in water, soluble detector. TSKgel G1000HXL, TSKgel G2000HXL, TSKgel in ethanol.” Following the descriptive terms for solubility G3000HXL and TSKgel G4000HXL columns (all 300×7.8 mm according to JECFA and shown in Table 2, solubility test- i.d., 5 µm) (Tosoh Corporation) were tandemly connected and ing in water was conducted at the lowest-limit concentration used for the separations. The SEFA samples were dissolved of “sparingly soluble” (1% (w/w)) and that in ethanol at the in tetrahydrofuran at a concentration of 5 mg/mL. The injec- lowest-limit concentration of “soluble” (3.3% (w/w)). tion volume was 80 µL, the flow rate was 0.8 mL/min, and the 1) Solubility Testing in Water mobile phase was HPLC-grade degassed tetrahydrofuran. The A sample of SEFA product (1 g) was placed in a 200-mL columns and refractive index detector temperatures were set conical flask, to which 100 g of water (at approximately 20°C) at 40°C. The percentage of monoester content in each sample was added. After sealing with Parafilm, the flask was shaken was calculated by the following formula: lightly to disperse the sample. Subsequently, the flask was placed in a 20°C water bath (constant temperature immersion- % sucrose monoester= 100 A /T type controller Thermomate BF-500, Yamato Scientific Co., where A is the peak area of the monoester and T is the sum of Ltd., Tokyo, Japan) and shaken for 5 min by hand. Solubility all peak areas eluting within 43 min of initiating elution. was assessed visually. 2) Analysis of Fatty Acid Composition 2) Solubility Testing in Ethanol The fatty acid composition of the 10 samples was analyzed A sample of SEFA product (1.6 g) was placed in a 100-mL by gas chromatography with flame ionization detection (GC- conical flask, to which 48 g of ethanol (at approximately 20°C) FID) according to the Standard Methods for the Analysis of was added. After sealing with Parafilm, the flask was shaken Fats, Oils, and Related Materials, as edited by the Japan Oil lightly to disperse the sample. Subsequently, the flask was Chemists’ Society.8) placed in a 20°C water bath and shaken for 5 min by hand. Fatty acid composition was analyzed on an Agilent 7890B Solubility was assessed visually. 286 Biol. Pharm. Bull. Vol. 40, No. 3 (2017)

Table 3. Monoester Content and Fatty Acid Composition of Sucrose Esters of Fatty Acids (SEFA) Product Samples

Fatty acid composition (%) Sample Fatty acid type Monoester content (%) C12 (Lauric) C16 (Palmitic) C18 (Stearic)

L-1695 C12 (Lauric) 80 99 Trace Trace L-595 C12 (Lauric) 35 99 Trace Trace P-1670 C16 (Palmitic) 78 N.D.* 79 21 S-1670 C18 (Stearic) 76 N.D.* 30 69 S-1170 C18 (Stearic) 59 N.D.* 29 71 S-570 C18 (Stearic) 30 N.D.* 30 70 F-110 C18 (Stearic) 57 N.D.* 28 71 F-50 C18 (Stearic) 40 N.D.* 28 71 11S C18 (Stearic) 56 N.D.* 39 59 5S C18 (Stearic) 27 N.D.* 31 69

* “N.D.” indicates “not detected.”

Effect of Heating on Solubility Solubility was also tested after samples were heated to increase the solubility. Table 4. Solubility in Water and Ethanol According to the JECFA Test 1) Solubility Testing in Water Methods A sample of SEFA product (1 g) was placed in a 200-mL Sample In water In ethanol conical flask, to which 100 g of water (at approximately 20°C) was added. After sealing with Parafilm, the flask was shaken L-1695 Transparent Transparent lightly to disperse the sample. Subsequently, the flask was L-595 White turbidity and precipitates Transparent placed in a 60°C water bath and shaken for 5 min by hand. P-1670 White turbidity and precipitates Transparent Solubility was assessed visually. S-1670 White turbidity and precipitates Transparent S-1170 White turbidity and precipitates Precipitates 2) Solubility Testing in Ethanol S-570 White turbidity and precipitates Precipitates A sample of SEFA product (1.6 g) was placed in a 100-mL F-110 White turbidity and precipitates Precipitates conical flask, to which 48 g of ethanol (at approximately 20°C) F-50 White turbidity and precipitates Precipitates was added. After sealing with Parafilm, the flask was shaken 11S White turbidity and precipitates Precipitates lightly to disperse the sample. Subsequently, the flask was 5S White turbidity and precipitates Precipitates placed in a 50°C water bath and shaken for 5 min by hand. Solubility was assessed visually. Measurement of Transmittance A sample of SEFA product (1 g) was placed in a 200-mL conical flask, to which 100 g of water (at approximately 20°C) was added. After sealing with Parafilm, the flask was shaken lightly to disperse the sample. Subsequently, the flask was placed in a 60°C water bath and shaken for 5 min by hand. The inside small bubbles resulting from shaking was removed by sonication for 10 min in an ultrasonic cleaner (2510J-MT, Yamato Scientific Co., Ltd.). Immediately after this step, the sample was placed in a 1-cm quartz cell, and the light transmittance was measured at a scan width of 0.2 nm over the range of 350 to 780 nm using a spectrophotometer (UV-1800, Shimadzu Corp., Kyoto, Japan).

RESULTS

Properties of Individual Samples To investigate the properties of each SEFA product sample, the monoester content was analyzed by gel permeation chromatography, and the fatty acid composition was analyzed by gas chromatography with flame ionization detection. The results are shown in Table 3. The lauric acid-type samples (L-1695 and L-595) consisted of nearly 100% lauric acid. As for the palmitic acid- and stearic acid-type samples, they consisted of a mixture of palmitic and stearic acid. The monoester content of these samples ranged from 27 to 80%. The monoester content of the Fig. 2. Visual Assessment of Solubility in Water and Ethanol Accord- L-1695, P-1670, and S-1670 samples was approximately 80%. ing to the JECFA Test Methods Solubility Testing The solubility of each SEFA sample To better assess solubility, the solutions were transferred from conical flasks to was evaluated according to the JECFA test methods. The centrifuge tubes, and photographs were taken. Vol. 40, No. 3 (2017) Biol. Pharm. Bull. 287 results are shown in Table 4. The results of solubility tests water in the JECFA specifications although warm water is of L-1695, P-1670, S-1170, and S-570 are also shown in pho- not defined in the JECFA test methods and procedures.11) In tographs (Fig. 2). Nine samples produced white turbidity and this study, we investigated their solubilities at 60°C in water precipitates in water, and 6 samples produced precipitates in because the temperature of warm water is defined as from 60 ethanol. Only the L-1695 sample dissolved transparently in to 70°C in the Japanese Pharmacopoeia. The results are shown both water (1% (w/w)) and ethanol (3.3% (w/w)) and met the in Table 5. The results of the solubility tests are also shown solubility criterion of the JECFA SEFA specifications (spar- in photographs (Figs. 3, 4). L-1695 dissolved transparently in ingly soluble in water, soluble in ethanol). P-1670 dissolved both water and ethanol, as expected. In ethanol, other samples transparently in ethanol but produced white turbidity in water; dissolved transparently. As shown in Fig. 3, in water, P-1670 thus, it did not meet the solubility criterion of the JECFA and S-1670 dispersed to assume a slightly white appearance, SEFA specifications. Only L-1695 met the solubility criterion whereas other samples, such as L-595, S-1170, S-570, F-110, of JECFA SEFA specifications. F-50, 11S, and 5S also assumed a white appearance with- Effect of Heating on Solubility We also investigated the out producing precipitates. Thus, the solubility of the SEFA effect of temperature on solubility, as the solubility of emulsi- samples increased following heating; however, none of the fiers is generally known to increase following heating.10) The samples except L-1695 dissolved transparently in water. melting points of the SEFA product samples ranged from 35 Measurement of Light Transmittance After the water to 50°C (data not shown). Thus, we investigated their solubili- solubility of heated samples was tested, the light transmit- ties at 50°C in ethanol in this study. As for the solubilities in tance of the sample solutions was measured. The results are water, there are some emulsifiers of which solubility criterion shown in Fig. 5. Samples that dissolved transparently or dis- in water is specified under heating conditions such as warm persed to assume a slightly white appearance, such as L-1695 and S-1670, showed more than 90% light transmittance in the wavelength range of 350 to 700 nm. Moreover, the light Table 5. Solubility in Warm Water (60°C) and Warm Ethanol (50°C) transmittance of L-1695 in the identical wavelength range Sample In water In ethanol was higher than that of S-1670. On the other hand, S-570 and S-1170 dispersed to assume a white appearance and showed L-1695 Transparent Transparent less than 35% and less than 90% light transmittance, respec- L-595 White dispersion Transparent tively, in this wavelength range. Thus, samples with a less P-1670 Slightly white dispersion Transparent turbid appearance showed higher light transmittance in the S-1670 Slightly white dispersion Transparent wavelength range of 350 to 700 nm. S-1170 White dispersion Transparent S-570 White dispersion Transparent DISCUSSION F-110 White dispersion Transparent F-50 White dispersion Transparent In this study, we investigated the solubility of 10 SEFA 11S White dispersion Transparent 5S White dispersion Transparent samples using the JECFA test methods and found that 9 of these samples produced white turbidity or precipitates and did

Fig. 3. Appearance of Samples in Warm Water (60°C) 288 Biol. Pharm. Bull. Vol. 40, No. 3 (2017)

Fig. 4. Appearance of Samples in Warm Ethanol (50°C)

Fig. 5. Light Transmittance of 4 Samples (S-570, S-1170, S-1670, and L-1695) in Warm Water (60°C) not dissolve transparently in water. The present study showed These results suggest that not only the content of monoesters that most of the SEFA products commercially available world- but also the type of fatty acids affects the solubility in water. wide do not meet the solubility criterion of the JECFA SEFA Although L-1695 meets the solubility criterion of the specifications. Although we do not know the details of the JECFA SEFA specifications, this product is rarely used as a process that the JECFA used to establish the solubility criteri- food additive because of its strong bitter taste. It is often used on as “sparingly soluble in water, soluble in ethanol,” we con- as a raw material in preparing antifogging agents for food sidered that very pure sucrose monoesters of fatty acids, such containers or detergents for washing foods. Generally, stearic as those used as reagents, had been used to establish the solu- acid-type SEFA products are used as food additives; however, bility criterion. Sample L-1695, which was the only sample they did not dissolve transparently and did not meet the solu- that dissolved transparently in both water and ethanol and met bility criterion of the JECFA SEFA specifications. Further- the solubility criterion of the JECFA SEFA specifications, has more, the stearic acid-type samples S-570 and S-1170, which a high content of lauric acid and high content of monoesters. also did not meet the solubility criterion of the JECFA SEFA However, S-1670 and P-1670, which also have a high monoes- specifications, were used as test materials in the safety studies ter content, produced white turbidity or precipitates in water. of SEFA evaluated by the JECFA. Accordingly, the current Vol. 40, No. 3 (2017) Biol. Pharm. Bull. 289

JECFA solubility specification for SEFA needs to be revised. Japan. As the JECFA specifications are international standards, it is extremely important that worldwide commercially available Conflict of Interest Chiye Tatebe, Kyoko Sato and Hi- food additives meet the specifications from the perspectives roshi Akiyama have no conflict of interest. Yukino Nagai, of quality and safety assurance. In addition, according to the Satoko Kawano and Masaaki Tomida are employees of the Agreement on the Application of Sanitary and Phytosanitary Mitsubishi Chemical Corporation, Chiyoda-ku, Tokyo, Japan. Measures (the SPS agreement), if an international trade dis- Kenichiro Motoda is an employee of the Mitsubishi-Kagaku pute would occur, judgment will most likely be made based on Foods Corporation, Yokohama, Japan. the international standard. Therefore, the fact that worldwide commercially available SEFA products do not meet the cur- REFERENCES rent solubility criterion of the JECFA SEFA specifications may cause international trade conflicts. 1) Yuki A, Matsuda K, Nishimura A. Effect of sucrose polyesters on In our present study, we found that the solubility of SEFA crystallization behavior of vegetable shortening and margarine fat. product samples in water or ethanol increased when the J. Jpn. Oil Chem. Soc., 39, 236–244 (1990). samples were heated. In water at 60°C, all of the samples dis- 2) Clark DC, Wilde PJ, Wilson DR, Wustneck R. The interaction of sucrose esters with β-lactoglobulin and β-casein from bovine milk. persed and became white without producing precipitates, and Food Hydrocoll., 6, 173–186 (1992). in ethanol at 50°C, all of the samples dissolved transparently. 3) Desay NB. Esters of sucrose and glucose as cosmetic materials. As We conclude that when the sample solutions were heated close renewable natural resources, carbohydrates are gaining importance to the melting points of the SEFA products, the solubility in- as chemical raw materials. Cosmetics and Toiletries, 105, 99–107 creased due to an increase in the molecular movement of the (1990). hydrophobic groups (i.e., the fatty acid moieties). In addition, 4) Schenk P, Ausborn M, Bendas F, Nuhn P, Arndt D, Meyer HW. The the results of measurement of light transmittance showed preparation and characterization of lipid vesicles containing esters that the appearance of the water-solubility at 60°C for SEFA of sucrose and fatty acids. J. Microencapsul., 6, 95–103 (1989). product samples is involved in the light transmittance in the 5) The Joint FAO/WHO Expert Committee on Food Additives. Com- wavelength range of 350 to 700 nm. pendium of Food Additive Specifications. 73rd Meeting 2010, Food and Agriculture Organization of the United Nations, Rome, pp. 23–30 (2010). CONCLUSION 6) The 44th Meeting of the Joint FAO/WHO Expert Committee on Food Additives. WHO Food Additives Series 35, World Health Or- Only 1 of 10 SEFA product samples tested in the pres- ganization, Geneva (1996). ent study met the solubility criterion of the JECFA SEFA 7) The 49th Meeting of the Joint FAO/WHO Expert Committee on specifications. The solubility improved upon heating; that is, Food Additives. WHO Food Additives Series 40, World Health Or- the heated samples dissolved transparently in ethanol and ganization, Geneva (1998). dispersed and became white without producing precipitates in 8) Standard Methods for the Analysis of Fats, Oils and Related Mate- water. This study suggests that the current solubility criterion rials, the Japan Oil Chemists’ Society, Tokyo (2013). of the JECFA SEFA specifications needs to be appropriately 9) The Joint FAO/WHO Expert Committee on Food Additives. Com- revised, as most of the SEFA products commercially available bined Compendium of Food Additive Specifications. Volume 4, Food and Agriculture Organization of the United Nations, Rome, p. worldwide do not meet the current solubility criterion of the 41 (2006). JECFA SEFA specifications. 10) Eastoe J. Surfactant Aggregation and Adsorption at Interfaces. Col- loid Science: Principles, Methods and Applications. (Cosgrove T Acknowledgments The authors are deeply grateful to Dr. ed.), Wiley, London, pp. 50–74 (2005). K. Seguro (Japan Food Additives Association) and DKS Co., 11) The Joint FAO/WHO Expert Committee on Food Additives. Com- Ltd. for their valuable comments in preparation of this paper. bined Compendium of Food Additive Specifications. Volume 4, This research was supported by a Grant-in-Aid for Scientific Food and Agriculture Organization of the United Nations, Rome Research from the Ministry of Health, Labour and Welfare of (2006).