483

Journal of Oleo ScienceCopyright ©2012 by Japan Oil Chemists’ SocietyJ. Oleo Sci. 61, (9) 483-489 (2012)

Preparing Glabridin-in-Water Nanoemulsions by High Pressure Homogenization with Response Surface MethodologyChang-Wei Hsieh1＊ , Po-Hsien Li2, I-Chi Lu3 and Teng-Hsu Wang4

1 Department of Medicinal Botany and Healthcare, Da-Yeh University (No.168, University Rd., Dacun, Changhua 51591, Taiwan, Republic of China)

2 Institute of Food Science and Technology, National Taiwan University, 1, Sec. 4, Roosevelt Road, Taipei, Taiwan, 10617, Republic of China)3 Department and Graduate Program of BioIndustry Technology, Da-Yeh University (No.168, University Rd., Dacun, Changhua 51591, Taiwan,

Republic of China)4 School of Pharmacy, National Defense Medical Center (No. 161, Sec. 6, Min-Chuan E. Rd., Taipei 104, Taiwan, Republic of China)

1 INTRODUCTIONGlabridin is a pyranoisoflavan isolated from licorice. It

has multiple pharmacological activities such as cytotoxic activity1） and antimicrobial activity against Helicobacter pylori2）, and methicillin-resistant Staphylococcus aureus3）. It has estrogenic and antiproliferative activity in human breast cancer cells4, 5）, and inhibits melanogenesis6）, inflammation7）, low-density lipoprotein oxidation8）, human cytochrome P450s 3A4, 2B6 and 2C9 activities9）, and ne-phritis10）. It protects mitochondrial functions against oxida-tive stress11）. Glabridin is the main compound in the hydro-phobic fraction of licorice extract but is susceptible to oxidative degradation, which results in low bioavailability and loss of pharmacological activities under normal storage conditions. It is also insoluble in water at neutral pH12, 13）. Therefore, emulsification technology maybe a good method to solubilize, encapsulate and protect this component.

Nanoemulsions are emulsion systems in the range of 50 to 200 nm. Because of their small droplet size, nanoemul-sions appear transparent or translucent to the naked eye and possess good stability against sedimentation, coales-cence, creaming and Ostwald ripening14－16）. The character-

＊Correspondence to: Chang-Wei Hsieh, Department of Medicinal Botany and Healthcare, Da-Yeh University, No.168, University Rd., Dacun, Changhua 51591, Taiwan, Republic of ChinaE-mail: welson6404@hotmail.com Accepted April 27, 2012 (received for review September 5, 2011)Journal of Oleo Science ISSN 1345-8957 print / ISSN 1347-3352 onlinehttp://www.jstage.jst.go.jp/browse/jos/　　http://mc.manusriptcentral.com/jjocs

istic properties of nanoemulsions are of interest for practi-cal application. Nanoemulsions are used in cosmetics as personal-care formulations, in agrochemicals for pesticide delivery. In the pharmaceutical field, they are used as drug delivery systems for parenteral, oral, ocular or transdermal administration17, 18）. Because nanoemulsions are non-equi-librium systems, they can be produced by high-energy emulsification, which provides energy in a short time and the largest homogeneous flow that can effectively reduce the droplet size15）.

High-pressure homogenization is extensively used in the food, pharmaceutical and biotechnology industries to emulsify, disperse, mix and process various products19）. In a high-pressure homogenizer, the oil and water mixture is subjected to intense turbulence and shear flow fields. Tur-bulence is said to be the predominant mechanism20）, al-though laminar shear and cavitation may also have impor-tant roles. Turbulence leads to the breakup of the dispersed phase into small droplets. The relative motion between the droplets results in their collision, thus leading to their coalescence. A dynamic equilibrium usually exists between breakage and coalescence. The shelf life of the

Abstract: Glabridin is a pharmacological active hydrophobic pyranoisoflavan isolated from licorice. It has low bioavailability and solubility and therefore is difficult to apply for industry use. We investigated the effect of combining caprylic triglyceride with glabridin (2-6%, w/w), emulsifier (3-7%, w/w), and homogenization pressure (70-130 MPa) on the droplet size of glabridin nanoemulsions using response surface methodology by a 3-factor-3-level Box-Behnken design. Oil content, emulsifier content and pressure had a significant effect on droplet size (p < 0.05). The optimal conditions for preparing glabridin nanoemulsions were predicted to be caprylic triglyceride (oil content), 3.7%; emulsifier content, 5.3%, and homogenization pressure, 129 MPa.

Key words: Glabridin, nanoemulsions, response surface methodology, high pressure homogenization

C.-W. Hsieh, P.-H. Li, I-C. Lu et al.

J. Oleo Sci. 61, (9) 483-489 (2012)

484

emulsion, as well as the texture, greatly depends on the droplet size distribution, which can be adjusted by control-ling the rate of droplet breakage and coalescence during emulsion formation. This control requires knowledge of the effect of operating parameters and formulation of the emulsion on droplet breakage and coalescence. Further-more, the emulsification conditions significantly affect the quality of the formed nanoemulsions and ought to be sys-tematically studied. Several publications have explored the parameters of emulsification on preparing nanoemulsions. For instance, β-carotene nanoemulsions was formed with 10％ Tween 20 by using high pressure homogenization21）, their results showed that parameters such as homogeniza-tion pressure can significantly affect droplet size and size distribution. Li and Chiang reported the preparation of D-limonene in water nanoemulsion by ultrasonic emulsifica-tion method using mixed emulsifiers of sorbitane trioleate and polyoxyethylene oleyl ether22）, and indicated that applied power, emulsification time and emulsifier concen-tration had significant effects on the droplet size of nano-emulsions. Liu et al. studied the paraffin oil in water nano-emulsions with mixed emulsifiers Tween 80/Span 80 by using the emulsion inversion point method at different emulsification temperatures23）. Their study showed that the emulsifier concentration significantly affects the droplet formation and stability of nanoemulsions.

We aimed to systematically investigate the effect of emulsifying conditions on the physicochemical properties of glabridin-in-water nanoemulsions with the smallest droplet size using response surface methodology（RSM）.

2 EXPERIMENTAL2.1 Materials

Polyoxythylene sorbitan monooleate（Tween 80）and sor-bitan monooleate（Span 80）was from Katayama Chemical Co.（Japan）, and were mixed at a ratio of 10:2（w/w）for the emulsifier phase. Ultrapure water was from Milli-Q Plus system（Molsheim, France）. Glabridin was from Propagate Trading Co.（Taipei）. The oil phase consisted of 3％ gla-bridin in capric triglyceride（First Chemical Works, Taiwan）.

2.2 Nanoemulsions preparationNanoemulsions consisted of the oil phase, mixed emulsi-

fier, and water phase. All emulsions were prepared in 2 stages. The coarse emulsions were prepared with use of a Polytron（PT-MR 3000, Kinematica AG, Littau, Switzer-land）, and then further emulsified by use of a high pressure homogenizer（EmulsiFlex-C3, AVESTIN inc., Ottawa, Canada）. During emulsification, the difference in tempera-ture from initial coarse emulsions to final emulsion was not more than 20℃. Each experiment was performed in tripli-cated.

2.3 Experimental design Response surface methodology by a 3-factor-3-level Box-

Behnken design was used to study the effect of the inde-pendent variables: oil phase concentration（X1）, emulsifier concentration（X2）and homogenization pressure（X3）, on droplet size（Y）of the nanoemulsions. The independent variables used in the RSM design are listed in Table 1. The experiments were designed according to the central com-posite design（CCD）with a 15 factorial and star design and 3 central points as shown in Table 2. Individual experi-ments were carried out in a random order. A second-order polynomial equation was used to express the droplet size（Y）of the nanoemulsions as a function of the independent variables as follows:

Y＝ a0＋a1X1＋a2X2＋a3X3＋a11X12＋a22X2

2＋a33X32＋

a12X1X2＋a13X1X3＋a23X2X3

where a0 is a constant, and ai, aii and aij are the linear, quadratic and interactive coefficients, respectively. The co-efficients of the response surface equation were deter-mined by use of Statgraphics Centurion XV（StatPoint, Inc., 2005）.

2.4 Droplet size analysis The droplet size of the nanoemulsions was determined

by use of the dynamic light scattering using a Zetasizer Nano-ZS90（Malvern Instruments, Worcestershire, UK）. The measurement was carried out at a fixed angle of 90° with the samples diluted approximately 1000 fold with use of Milli-Q water. Emulsion droplet size was estimated as the mean diameter of the volume distribution（MV）from 3 measurements.

Table 1　Independent variables used in RSM design.

Independent variable SymbolCoded variables

－ 1 0 1

Oil phase concentration (%) X1 2 4 6

Surfactant concentration (%) X2 3 5 7

Homogenization pressure (MPa) X3 70 100 130

Preparation of Glabridin-in-Water Nanoemulsions

J. Oleo Sci. 61, (9) 483-489 (2012)

485

MV＝ ΣVidi

ΣVi

where Vi is the volume fraction between droplet sizes and di is the diameter of droplets.

2.5 Transmission electron microscopic analysisThe morphologic features of the glabridin nanoemulsions

were visualized by transmission electron microscope（TEM）. Samples（50 μL）were added to 200-mesh formwar-coated copper TEM sample holders（EM Sciences, Hatfield, PA）, then negatively stained with 50 μL of 1.5％（w/v）phosphotungstic acid for 10 min at room temperature. Excess liquid was blotted with a piece of Whatman filter paper. The TEM samples were observed with use of the JEOL JSM-1200EX II transmission electron microscope（Peabody, MA）with a 20 μm aperture at 67 kV.

3 RESULTS AND DISCUSSION3.1 Analysis of response surfaces

The droplet size of the glabridin nanoemulsions obtained from all the experiments are in Table 2. The experimental data was used to calculate the coefficients of the quadratic polynomial equation, and the derived equation was then used to predict the values of droplet size of the nanoemul-sions. The predicted values agreed well with the experi-mental values obtained from the RSM design. ANOVA showed that the resulting quadratic polynomial models ad-equately represented the experimental data with a coeffi-cient of multiple determinations（R2）of 0.945.

The coefficient estimates and the corresponding P-values for the polynomial equation suggested that all of 3 inde-pendent variables had a significant effect on the formation of glabridin nanoemulsions（P＜0.05）（Table 3）. The qua-

Table 2　The experimental droplet size and predict droplet size obtained from central composite design.

Experimental numbera

Oil content(%)X1

b

Surfactant content (%)

X2b

Pressure (MPa)

X3b

Experimental size (nm)

Y1c

Predicted size (nm)

Y2

1 2 (－1) 3 (－1) 100 (0) 87±3 85.05

2 6 (1) 3 (－1) 100 (0) 110±4 115.68

3 2 (－1) 7 (1) 100 (0) 87±4 81.43

4 6 (1) 7 (1) 100 (0) 73±1 76.05

5 2 (－1) 5 (0) 70 (－1) 104±7 113.31

6 6 (1) 5 (0) 70 (－1) 122±3 123.39

7 2 (－1) 5 (0) 130 (1) 52±4 51.61

8 6 (1) 5 (0) 130 (1) 75±5 66.79

9 4 (0) 3 (－1) 70 (－1) 138±3 132.04

10 4 (0) 7 (1) 70 (－1) 103±2 101.06

11 4 (0) 3 (－1) 130 (1) 60±4 63.54

12 4 (0) 7 (1) 130 (1) 45±7 51.26

13 4 (0) 5 (0) 100 (0) 56±2 56.1

14 4 (0) 5 (0) 100 (0) 57±1 56.1

15 4 (0) 5 (0) 100 (0) 56±2 56.1a The experiments were run in random order.b The values (－1), (0), and (1) are independent variables used in response surface methodology design.c The values are the mean and standard deviation (n=3).

Table 3　��Analysis by ANOVA of the regression coefficients of the quadratic equation on droplet size of glabridin nanoemulsions.

Source Degrees of freedom Sum of squares F-Value P-value

Linear 3 8251.51 43.56 0.0005

Quadratic 3 2420.63 12.78 0.0080

Cross product 3 417.93 2.21 0.2055

Total model 9 11090 19.52 0.0022

Lack of fit 3 314.69 212.63 0.0047

C.-W. Hsieh, P.-H. Li, I-C. Lu et al.

J. Oleo Sci. 61, (9) 483-489 (2012)

486

dratic term also had significant effect on the droplet size（P＜0.05）. Therefore, the following second-order polynomial equation was used to predict the droplet size of glabridin nanoemulsions: Y＝501.72269－22.96875X1－42.92708X2

－0.47308X3＋4.32292X12＋3.87292X2

2＋0.00016352X32

3.2 Optimization of conditions for preparing nanoemul-sions

Response surface methodology is an empirical modeling tool consisting of a group of mathematical and statistical techniques that can be used to develop, improve and opti-mize processes in which the response is influenced by several variables24）. RSM allows researchers to conduct op-timization study with ease as it helps to reduce the number of experimental trials required to as minimum as possible. We generated a surface response of the quadric polynomial model by varying 3 of the independent variables within the experimental range while holding another one constant at the central point. Thus, Fig. 1 was generated by varying the applied pressure and emulsifier content while holding oil content at 4％. Homogenization pressure can signifi-cantly influence the properties of emulsions, because pres-sure dependency of the shear forces and turbulence pro-duced during homogenization, both pressure dependent, can affect the droplet size and size distribution. In this study, increasing the homogenization pressure resulted in a significant decrease in droplet sizes over the entire pres-sure ranges studied, which agreed with previous studies19－23）. The initial average radius of the droplets declines from about 130 to 60 nm as the emulsifier concentration ranged from 3.0 to 5.0 wt％. Increasing the emulsifier concentra-

tion from 5.5 to 7.0 wt％ further lowers the droplet size to less than 50 nm in diameters. This can be explained by the added amount of emulsifier increases the total interfacial area and hence enables the stabilization of emulsion drop-lets at lower sizes25）.

Figure 2 was generated by varying the applied pressure and oil content while maintaining the emulsifier content at 5％. In general, the droplet size increased with increasing concentration of oil. Among the pressure range provided, a desirable threshold oil content that produces lower range of droplet size was shown at range between 3.0 and 4.0 ％. During high pressure homogenization emulsification, in-creasing of oil concentration can result in the increasing of amount of newly formed droplet, as well as increased extent of coalescence26）. It is likely that once the threshold oil content is reached, the additional oil phase promotes the disruption of interfacial layer that initially stabilized by limited emulsifiers.

Figure 3 was generated by varying the emulsifier and oil content while maintaining the applied pressure at 100 MPa. The droplet size increased with increasing of concentration of emulsifier, in that the newly formed droplet had suffi-cient emulsifier to stabilize the emulsion system. Our result agreed with those from previous studies27－29）. When the concentration of emulsifier was＜4.5％, the size of the droplet increased. If the emulsifier concentration is low and the interfacial area generated during emulsification is large, a considerable amount of droplets will not have enough emulsifier to overcome coalescence. After emulsification, droplets will coalesce until the total interfa-

Fig. 1　�Response surface contour plot of the combined effects of pressure, emulsifier content and oil content on the formation of glabridin nanoemul-sions: applied pressure and emulsifier content was processed at 4% oil contents.

Fig. 2　�Response surface contour plot of the combined effects of pressure, emulsifier content and oil content on the formation of glabridin nanoemul-sions: applied pressure and oil content were pro-cessed at 5% emulsifier contents.

Preparation of Glabridin-in-Water Nanoemulsions

J. Oleo Sci. 61, (9) 483-489 (2012)

487

cial area of the emulsion is stabilized. The final size of the droplet is a collective result of the emulsifier adsorption time and the emulsifier concentration of the adjacent liquid30, 31）.

The condition for the preparing glabridin nanoemulsions would be optimal with the smallest size of nanoemulsions droplets. Therefore, from the results we described above, we chose the oil and emulsifier contents of 3.7％ and 5.3％, respectively, as the optimal conditions, under homogeniza-tion pressure of 129 MPa. The droplet size distribution in the glabridin nanoemulsions at optimal conditions is in Fig. 4. The droplet size distributions were unimodal and typi-cally extended from 20 to 120 nm. Therefore, high-pres-sure homogenization emulsification can be used to produce glabridin nanoemulsions with diameter in the nanometer range by RSM.

3.3 Appearance of glabridin nanoemulsionsThe glabridin nanoemulsions were translucent to the

naked eyes. TEM was used to observe glabridin droplets in the nanoemulsions. Figure 5 shows phosphotungstic acid stained glabridin droplets visible by TEM, and the droplet size agreed with the results of droplet size analysis. In ad-dition, TEM images showed discrete dark spherical outline with monodispersed size distribution of the glabridin nano-emulsions. Due to the advantages of extremely small droplet sizes, translucent appearance and very large dis-persed phase surface-to-volume ratios, glabridin nanoemul-sions should be attractive for cosmetics, pharmaceutical

and food industries, as they may have the potentials to be readily absorbed by the skin, easily sterilized by filtration, and can deliver high concentrations of glabridin to the targets of interests. Further investigations are currently undergoing to elucidate the applications of glabridin nano-emulsions.

4 CONCLUSIONUse of RSM with a second-order polynomial model was

sufficient to describe and predict the response of the droplet of glabridin nanoemulsions. The independent vari-ables of homogenization pressure, oil and emulsifier content and the quadrics of the 3 variables had a significant effect on the droplet size of glabirdin nanoemulsions. We adopted an optimization method to determine the best emulsifying conditions. The optimal conditions for prepar-ing glabridin nanoemulsions were predicted to be emulsifi-er（Tween 80＋Span 80）concentration 5.3％, caprylic tri-glycerides concentration 3.65％, and homogenization

Fig. 3　�Response surface contour plot of the combined effects of pressure, emulsifier content and oil content on the formation of glabridin nanoemul-sions: emulsifier content and oil content were processed at 100 MPa pressures.

Fig. 4　�The droplet size distribution of glabridin nano-emulsions with oil content 3.65%, emulsifier content 5.3% and 129 MPa pressure.

Fig. 5　�Transmission electron micrograph of droplets in the glabridin/water nanoemulsion system.

C.-W. Hsieh, P.-H. Li, I-C. Lu et al.

J. Oleo Sci. 61, (9) 483-489 (2012)

488

pressure 129 MPa.

5 AcknowledgementThe authors are very grateful to Morita Biotech Corpora-

tion for the financial support on the present study.

References1） Fukai, T.; Sakagami, H.; Toguchi, M.; Takayama, F.;

Iwakura, I.; Atsumi, T; Ueha, T.; Nakashima, H.; Nomu-ra, T. Cytotoxic activity of low molecular weight poly-phenols against human oral tumor cell lines. Antican-cer Res. 20, 2525-2536.（2000）.

2） Fukai, T.; Marumo, A.; Kaitou, K.; Kanda, T.; Terada, S.; Nomura, T. Anti-helicobacter pylori flavonoids from licorice extract. Life Sci. 71, 1449-1463（2002）.

3） Hatano, T.; Shintani, Y.; Aga, Y.; Shiota, S.; Tsuchiya, T.; Yoshida, T. Phenolic constituents of licorice. VIII. Structures of glicophenone and glicoisoflavanone, and effects of licorice phenolics on methicillin-resistant Staphylococcus aureus. Chem. Pharm. Bull 37, 1286-1292（2000）.

4） Tamir, S.; Eizenberg, M.; Somjen, D.; Stern, N.; Shelach, R.; Kaye, A.; Vaya, J. Estrogenic and antipro-liferative properties of glabridin from licorice in hu-man breast cancer cells. Cancer Res. 60, 5704-5709（2000）.

5） Somjen, D.; Knoll, E.; Vaya, J.; Stern, N.; Tamir, S. Es-trogen-like activity of licorice root constituents: gla-bridin and glabrene, in vascular tissues in vitro and in vivo. J. Steroid. Biochem. Mol. Biol. 91, 147-155（2004）.

6） Yokota, T.; Nishio, H.; Kubota, Y.; Mizoguchi, M. The inhibitory effect of glabridin from licorice extracts on melanogenesis and inflammation. Pigment Cell Res. 11, 355-361（1998）.

7） Nerya, O.; Vaya, J.; Musa, R.; Izrael, S.; Ben-Arie, R.; Tamir, S. Glabrene and isoliquiritigenin as tyrosinase inhibitors from licorice roots. J. Agric. Food Chem. 51, 1201-1207（2003）.

8） Rosenblat, M.; Belinky, P.; Vaya, J.; Levy, R.; Hayek T.; Colemani, R.; Merchav, S.; Aviram, M. Macrophage en-richment with the isoflavan glabridin inhibits NADPH oxidase-induced cell-mediated oxidation of low densi-ty lipoprotein. J. Biol. Chem. 274, 13790-13799（1999）.

9） Kent, U. M.; Aviram, M.; Rosenblat, M.; Hollenberg, P. F. The licorice root derived isoflavan glabridin inhibits the activities of human cytochrome P450S 3A4, 2B6, and 2C9. Drug Metab. Dispos. 30, 709-715（2002）.

10） Fukai, T.; Satoh, L.; Nomura, T.; Sakagami, H. Prelimi-

nary evaluation of antinephritis and radical scavenging activities of glabridin from Glycyrrhiza glabra. Fito-terapia 74, 624-629（2003）.

11） Haraguchi, H.; Yoshida, N.; Ishikawa, H.; Tamura, Y.; Mizutani, K.; Kinoshita, T. Protection of mitochondrial functions against oxidative stresses by isoflavans from Glycyrrhiza glabra. J. Pharm. Pharmacol. 52, 219-223（2000）.

12） Ao, M.; Shi, Y.; Cui, Y.; Guo, W.; Wang, J.; Yu, L. Factors influencing glabridin stability. Nat. Prod. Commum. 5, 1907-1912（2010）.

13） Mambro, V. M. D.; Fonseca, M. J. V. Assays of physical stability and antioxidant activity of a topical formula-tion added with different plant extracts. J. Pharma-ceut. Biomed. 37, 287-295（2005）.

14） Tadros, T.; Izquierdo, P.; Esquena, J.; Solans, C. For-mation and stability of nano-emulsions. Adv. Colloid Interfac. 108-108, 303-318（2004）.

15） Solans, C.; Izquierdo, P.; Nolla, J.; Azemar, N.; Garcia-Celma, M. J. Nanoemulsions. Curr. Opin. Colloid In. 10, 102-110（2005）.

16） Pey, C. M.; Maestro, A.; Solé, I.; González, C.; Solans, C.; Gutiérrez, J. M. Optimization of nano-emulsions pre-pared by low-energy emulsification methods at con-stant temperature using a factorial design study. Col-loid Surface A 288, 144-150（2006）.

17） Sonneville-Aubrun, O.; Simonnet, J. T.; L’Alloret, F. Nanoemulsions: a new vehicle for skincare products. Adv. Colloid Interfac. 108-109, 145-149（2004）.

18） Lijuan, W.; Xuefeng, Li.; Gaoyong, Z.; Jinfeng, D.; Ju-lian, E. Oil-in-water nanoemulsions for pesticide for-mulations. J. Colloid Interf. Sci. 314, 230-235（2007）.

19） Floury, J.; Desrumaux, A.; Lardières, J. Effect of high-pressure homogenization on droplet size distributions and rheological properties of model oil-in-water emul-sions. Innov. Food Sci. Emerg. 1, 127-134（2000）.

20） Walstra, P. Effect of homogenization on the fat globule size distribution in milk. Neth. Milk Dairy J. 29, 279-294（1975）.

21） Yuan, Y.; Gao, Y.; Mao, L.; Zhao, J. Optimisation of con-ditions for the preparation of β-carotene nanoemul-sions using response surface methodology. Food Chem. 107, 1300-1306（2008）.

22） Li, P. H.; Chiang, B. H. Process optimization and stabil-ity of D-limonene-in-water nanoemulsions prepared by ultrasonic emulsification using response surface meth-odology. Ultrason. Sonochem.19, 192-197（2012）.

23） Liu, W.; Sun, D.; Li, C.; Liu, Q.; Xu, J. Formation and stability of paraffin oil-in-water nano-emulsions pre-pared by the emulsion inversion point method. J. Col-loid Interf. Sci. 303, 557-563（2006）.

24） Bas, D., Boyac, I. H. Modelling and optimization I: us-ability of response surface methodology, J. Food Eng. 78, 836-845（2007）.

Preparation of Glabridin-in-Water Nanoemulsions

J. Oleo Sci. 61, (9) 483-489 (2012)

489

25） Gutiérreza, J. M., Gonzáleza, C., Maestroa, A., Solèa, I., Peya, C. M., Nollab, J. Nano-emulsions: New applica-tions and optimization of their preparation. Curr Opin Colloid Interf. Sci. 13, 245-251（2008）.

26） Jafari, S. M.; He, Y.; Bhandari, B. Production of sub-mi-cron emulsions by ultrasound and microfluidization techniques. J. Food Eng. 82, 478-488（2007）.

27） Bouchemal, K.; Briançon, S.; Perrier, E.; Fessi, H. Na-no-emulsion formulation using spontaneous emulsifi-cation: solvent, oil and surfactant optimization. Int. J. Pharm. 280, 241-251（2004）.

28） Jafari, S. M.; He, Y.; Bhandari, B. Optimization of nano-emulsions production by microfluidization. Eur. Food

Res. Technol. 225, 733-741（2007）.29） Henry, J. V. L.; Fryer, P. T.; Frith, W. J.; Norton, L. T.

Emuldification mechanisms and storage instability of hydrocarbon-in-water sub-micron emulsion stabilized with Tweens（20 and 80）, Brij 96 and sucrose monoes-ters. J. Colloid Interf. Sci. 338, 201-206（2010）.

30） Ivanov, I. B.; Danov, K. D.; Kralchevsky, P. A. Floccula-tion and coalescence of micron-size emulsion droplets. Colloid Surface A 152, 161-182（1999）.

31） Urbina-Villalba, G.; Garcia-Sucre, M. Influence of sur-factant distribution the stability of oil/water emulsions towards flocculation and coalescence. Colloid Surface A 190, 111-116（2001）.