The multiple emulsions

Introduction:

Seifriz started his pioneering work about multiple emulsions since 1925, which is regarded as the fundamental knowledge in the later research. Multiple emulsions are complicated systems which are considered as “emulsions of emulsions” (Garti, 1996).In the outer continues phase, the droplets of the dispersed phase named as globules which contain even smaller dispersed droplets ,the globules are separated from each other in external continues phase by a layer of oil phase film. In the inner phase, the droplets are departed from each other by oil phase (Benichou et al. 2006). It is widely believed that there exist two primary types of multiple emulsions, one is water-in-oil-in-water (W1/O/W2) emulsions that an w/o emulsion is dispersed in another aqueous phase (W2) and the other is oil-in-water-in-oil (O1/W/O2) emulsions that an o/w emulsion is dispersed in another oil phase(O2). In the previous study, water-in-oil-in-water (W1/O/W2) multiple emulsions have accounted for a vital role in the research of multiple emulsions , because the applications of W1/O/W2 multiple emulsions plays an important role in the food industry and it is also easier for us to select various of hydrophilic emulsifiers which are safe to health as stabilizers in preparation of multiple emulsions (Pays et al., 2002). As shown in Fig. 1, take water-in-oil-in-water (W1/O/W2) double emulsions as an example, which are composed of three distinct phases : an internal aqueous phase (W1), which containing many aqueous soluble ingredients. Various internal aqueous droplets are encapsulated in an oil phase (O), which is included in external aqueous phase (W2) (Garti, 1996).

Applications of multiple emulsions

It is widely believed that the potential applications so numerous that the research in such promising area can bring beneficial effects, especially in products’ areas ,such as drug-delivery systems, cosmetics, and foods . Water-in-oil-in-water (W1/O/W2) emulsions allow the encapsulation of active ingredients which have the ability to be soluble in the internal aqueous phase, thus it is possible to hide smell of some matter; remove toxic substance; or select appropriate conditions to realize controlled release of the active ingredients under certain process of emulsification. (Kanouni et al. 2002) On the basic of slow and sustained release of active ingredients from an internal reservoir into the external aqueous phase, the main function of double emulsions is regarded as an internal reservoir to entrap ingredients whatever you choose into the inner confined space, in order to protect against oxidation, light and enzymatic degradation. As a result, sensitive and active molecules can be protected from the external phase by the function of internal reservoir. In addition, because of the phenomenon of release of water or ingredients which can be observed in the experiments, the active ingredients will exist partly in the internal aqueous phase, partly in the oil phase and occasionally in the external phase(Benichou et al. 2004) .In the food industry, double emulsions provide some advantages because of their capability to encapsulate some water-soluble substances, such as flavours or active ingredients which are then slowly released from the internal compartments. Additionally, we should select food-grade additives which is soluble in the internal aqueous phase because the consumer products in food industry will be applied in our daily lives. Furthermore, as the development of needs in food quality, the production of low calorie and reduced fat products come into food market. (Muschiolik, 2007; Van der Graaf et al., 2005). In agrochemical industry, it has become increasingly difficult for scientists to produce products, such as pesticides which are effectively and simultaneously friendly to the environment. According to ElShafei et al. (2009), the idea of multiple emulsions has been successfully applied to the agriculture products and the multiple emulsions are relatively stable even on storage at room temperature and 4 ?for 30 days. As government increasingly pay attention to the safe and environmentally friendly products, the research in this orientation has draw public’s attention. Till now, no pharmaceutical multiple emulsions

have been brought to the market, because potential emulsifiers used in multiple emulsions are only available in cosmetic grade but not be applied in pharmaceutical grade. (Schmidts et al., 2009 ) In cosmetic area, the possibility of combining incompatible substances in products in order to offer more favorable functions. (Vasiljevic et al., 2005) multiple emulsions also have the potential to change the commonly oily feel of hand-cream to aqueous texture. The advance of products of cosmetics has brought out more space to develop in order to get more profits. (Kanouni et al., 2002)

Methods of preparation:

Scientists have done some research in multiple emulsions as the applications provide us more convenience and bring better consumer products in many areas. Because double emulsions have more complex structure and are even more thermodynamically unstable than single emulsions, they prone to be difficult to prepare, especially on an industrial scale. The difficulties of preparation of multiple emulsions have draw scientists’ attention, so many research have been pour into this area. In general, there exist single -step and two -step emulsification methods to prepare multiple emulsions (Allouche et al., 2003). Due to a multiple emulsion is considered as a mesophase between O/W and W/O emulsion, the one-step method of preparation means a combination of the two different types of emulsions and surfactant phase, which is very difficult to control. So, such method will not be chosen in the preparation (Matsumoto, 1987; Mulley and marland, 1980).

On the basic of previous study, the two-step emulsification process is considered as the most common and better controlled method. First of all, W1/O emulsions are much easier to prepare and it is also easy to control various characteristics in these emulsions as the parameters in them are relatively limited . Secondly, in the second step, it is widely believed that the complex structure and variable quantities result in relatively difficult to control or regulate. Many methods have been commonly used to improve the preparation of multiple emulsions, adding suitable emulsifiers is regarded as one of the most significant one.

In general, two kinds of emulsifiers are introduced to add in the preparation of multiple emulsions as the difference of their functions. Because of the different affinity of the emulsifiers, hydrophobic emulsifier ‘Emulsifier I’ which is used in the oil phase and hydrophilic emulsifier ‘ Emulsifier II’ which is used in the external aqueous phase (Garti, 1996). The hydrophobic emulsifier is designed to stabilize the interface of the W1/O internal emulsion and the hydrophilic emulsifier acts as stabilizer at the external interface of W1/O/W2 emulsion. The main function of emulsifiers is enhancing the stability of multiple emulsions in the preparation and even the long-time storage. The process of two-step preparation is shown in Fig.2. In the first step, the primary W/O emulsion is prepared under high-shear conditions (homogenization) to obtain small droplets, whereas the second step is carried out with less shear in order to avoid rupturing the internal droplets because the second step is much difficult to control than the first step (van der Graaf et al., 2004).

On the basic of Kanouni et al., (2002)’s earlier work, in the first step, they usually use an Ultra-Turrax mixer with a relatively high speed to prepare

a W1/O emulsion which is a combination of internal aqueous phase and an appropriate oil phase with suitable low HLB emulsifier; in the second step, the W1/O/W2 multiple emulsions will be produced by adding proper high HLB emulsifiers using Ultra-Turrax mixer or mechanical agitator with relatively smaller rotation speed.

In the previous study, stirring apparatuses, rotor-stator systems and high pressure homogenizers are considered as the most commonly and conventional emulsification devices (Schubert and Armbruster, 1992).As shown in table 1. the functions and disadvantages has been tabulated.

There are several drawbacks in such existing methods of production ( Williams et al.,1998). First of all, it is not easy for us to control the droplet size and droplet size distribution of the final multiple emulsions products. Secondly, it is difficult to scale up because different classes of the products are generated per batch on the same manufacture conditions, which contribute to one of the main factors why such products can not be applied in the industry. Moreover,van der Graaf et al. (2005) illustrate that conventional methods are not feasible in preparation of double emulsions, because high-shear stresses can result in rupture of the internal emulsions which should be avoided in the secondary emulsification (van der Graaf et al., 2005)

Different kinds of emulsification devices can generate various multiple emulsions with different conditions, such as droplet size, encapsulation efficiency, release rate, and so on. What has interested the scientists most recently is researching novel approaches to improve the emulsification equipment in order to generate more stable and ideal multiple emulsions. Much attention has been put in the improvement of the second step by using various pieces of equipment and novel method. Nakashima et al. (1991) points out that membrane emulsification is widely accepted as one of the new method for the production of emulsions recently( Nakashima et al., 1991). This technique is increasingly attracted because of its low energy consumption, the better control of droplet size and droplet size distribution and especially the mildness of the process, especially suitable to be used in the second step to prevent rupture of the double emulsion droplets (van der Graaf et al., 2005).

Joscelyne and Tragardh (1998) demonstrate that it is favourable to prepare small droplets when the conditions are higher concentrations of emulsifiers, high wall shear stress through a membrane with small pore size. As shown in Fig.3. because of the mild conditions in the process of membrane emulsification, it is easier to produce small size droplets and protect the multiple emulsions from membrane rupture, especially useful in the second step of emulsification. The system chosen ceramic membranes of different average pore size to prepare relative small droplets in multiple emulsions because such kinds of emulsions more stable. Membrane technology can be applied to the many productions, such as oil-in-water (O/W) emulsions t, UHT products and so on (Joscelyne and Tragardh 1998) .However, low flux of the dispersed phase is the main and visible drawback of membrane emulsification (Charcosset et al., 2004),which is caused by the properties of membranes with a low hydraulic . In general, two methods are commonly introduced in membrane emulsification: cross-flow membrane emulsification and pre-mix membrane emulsification (Suzuki et al., 1998). Take pre-mix membrane emulsification as an example, as shown in Fig.4. the most significant advantages of such method is it can provide high flux, which can improve the membrane emulsification process.

Various novel methods have been reported to improve the disadvantage of membrane emulsification. (Gijsbertsen-Abrahamse et al., 2004) for example, with the advance in nano- and micro engineering, it is possible to produce membranes with a low hydraulic resistance named microsieves. (Van Rijn et al., 2005) Microsieves, inorganic membranes, which can offer a very thin selective layer, high controlled pore size and shape, and smooth surfaces. As shown in Fig.5., SEM images of pore morphology of a silicon nitride microsieve surface. Microsieve membranes contribute to flux decline in crossflow filtration of bovine serum albumin (BSA) solutions. (Giron`es et al.,2006)

According to Shnji Sugiura et al., (2003), monodispersed multiple emulsions which are good at providing relatively stable conditions are regularly applied in industries and basic studies, on the basic of easier observation, monodispersed emulsions are regarded as an effective approach in determining the resistance to coalescence of an emulsion, and in observing how the active matter go through the oil film by diffusion. (Sugiura et al., 2003) Furthermore, a microfabricated channel array has been pointed out as a promising method for preparing monodisperse emulsion droplets (Kawakatsu et al., 1997). This type of emulsification technique is called microchannel (MC) emulsification, which is regarded as a novel method for preparing monodisperse emulsions. Owning to the advantages of this technique, it is a promising technique to improve the stability of multiple emulsions. (Kawakatsu et al., 2001; Sugiura et al., 2001 ). Nakagawa et al.(2004) suggest that monodisperse surfactant-free microcapsules can be produced by MC emulsification using gelatin. Of course, this technique need further study to improve its low production rate.

Improvements in stability of multiple emulsions

In practice, significant problems may arise, not only the thermodynamic instability of emulsions, but also many destabilization phenomenon, such as flocculation, coalescence and creaming, have contribute to the unstable emulsions (Vasiljevic et al., 2005). In order to protect the emulsions from the formation of flocculation or coalescence, two methods have been introduced to protect the droplets from each other, one is increasing viscosity of the external phase, the other is energy barrier. The DLVO theory is commonly applied to explain colloidal stability. when the distance between two colloid particles is increasing from small to large, the resulting potential is rage from negative to positive because the existence of attraction potential and repulsion potential ( Friberg, 1997).

Various factors may have an effect on the stability of multiple emulsions, including the method of preparation, the oil type, type and concentration of the emulsifier and so on (Vasiljevic et ,al. 2005). On the basic of fundamentally experimental data, we choose the relatively suitable and effective conditions to prepare multiple emulsions.

Many research have been put into how to improve the stability of multiple emulsions because thermodynamically unstable multiple emulsions not only exist in the process of preparation ,but also occur during storage or on exposure to environmental stresses such as mechanical forces, thermal processing, freezing or dehydration. On the basic of developed techniques, we can observe or measure the leakage of the inner aqueous phase(W1) in the outer phase and destabilization properties of the emulsions. There are four mechanisms explaining the instability of W1/O/W2 multiple emulsions: (1) the instability comes from the inner aqueous droplets because of coalescence; (2) the instability comes from the oil droplets because of coalescence; (3) rupture of the oil film (4) transport of water and ingredients through the oil layer (Appelqvist et al., 2007,; Florence and Whithill,1981; der Graaf et al., 2005).

In the real conditions, there may exist more than one mechanism in the multiple emulsions, different results to different situations. The determining of primary mechanisms exist in certain multiple emulsions should dependent on the experimental data and convincing analysis. What should we do is research more reasonable methods to solve the problem of thermodynamically unstablity in multiple emulsions. Three kinds of approach aiming at improving stabilization and slow solute release have been list as follows (Davis et al., 1985) : (1) stabilization of the inner W1/O emulsion, for example, the addition of various emulsifier combinations (Apenten and Zhu, 1996; Shima et al., 2004; Su et al., 2006); (2) stabilization of oil phase by choosing suitable oil type and the addition of proper carriers, complexants and viscosity builders, for instance, the solidification of the oil phase and the modification of the solubility and polarity of the oil phase to make it less water soluble (Tedajo et al., 2001); (3) stabilization of the external aqueous phase, such as increasing the viscosity of the outer aqueous phase (-zer, et al., 2000). Although many strategies have been categorized above, a majority of them are not suitable to apply in food industry because they are not easily scaled up in industry or they include not food- grade ingredients entrapped in multiple emulsions, which may make a bad influence on human health. So, there exists numerous space for us to research in the methods of improving the stability of multiple emulsions. (O’Regan and Mulvihill, 2009)

In general, many factors contribute to the improvement of stability of multiple emulsions as some research have deeply determined the main causes of thermal unstable phenomenon and flocculation, coalescence and creaming phenomenon. The nature and internal properties of surfactants or emulsifiers play a vital role in solving problem.

Stability of multiple (Opawale, et al., 1998) emulsions has been shown to be dependent on emulsifier interfacial film strength, ionic strength, various additives, and concentration. According to Vasiljevic et al. (2005), when the concentration of emulsifier in oil phase is higher, the multiple emulsions will have lower droplet size, higher viscosity and elastic characteristics. Moreover, changing the concentration of surfactants, results in the difference of the amount of retinol released from silica particles. In addition, different polymers which are added into the aqueous phase, the encapsulation efficiency of retinol was also changed (Hwang et al., 2005). The process of multiple emulsion formation and various destabilization processes can be determined by video microscopy (Ficheux et al., 1998). A unique dimpled structure is a signal to show the deformation of the multiple droplets and coalescence of the internal dispersed phase by coverslip pressure. If the multiple emulsions possess relatively high stability, then such structure come out for long-time observed in the presence of adequate concentrations of surfactants and additives. So, Formation of the dimple structure is linked with interfacial film strength and long-term multiple emulsion stability (Jiao et al., 2002).

The long-term stability of the double emulsion requires a balance between the Laplace and osmotic pressures among droplets in W1, because a stable W1/O emulsion is a fundamental and significant step in order to prepare a stable W1/O/W2 double emulsion.

Garti (1996) illustrate the concept of ‘weighted hydrophile-lipophile balance (HLB)’ is important because the value is linked with the droplet size, the number of W1 dispersed in inner phase and the stability of the

W/O/W multiple emulsions. Such properties are so significant in preparing relatively stable multiple emulsions that the weighted HLB value is considered as a potential reference to select the optimal type of emulsifiers in forming multiple emulsions.

In the first step of preparation, HLB(I) stands for the HLB value of the hydrophobic emulsifier, CI means the weight percentage of the hydrophobic emulsifier in the fundamental W1/O emulsion, In the second step of preparation, HLB(II) stands for the HLB value of the hydrophilic emulsifier, and CII means the weight percentage of the hydrophilic emulsifier in the W1/O/W2 multiple emulsion

It was observed that using a combination of an amphoteric high HLB surfactant and an anionic surfactant can prepare a stable system(Kanouni et al., 2002). The inner phase is demonstrated to be better stabilized by minimizing the size of droplets and forming microemulsion droplets or microsphere particles, or applying more potential surfactants in order to seal the active ingredients in the interface (O’Regan and Mulvihill, 2009). Choosing of optimal surfactants has made a positive effect on controlling particle size in multiple emulsions. Sepideh Khoee and Morteza Yaghoobian (2008) propose that the mean diameters of nanocapsules containing penicillin-G are linked with the properties of surfactants. that is to say, the different types or content of surfactant used in formation of multiple emulsions can result in different droplets’ size. N. Heldt et al. (2000) point put that changing the ratio of lecithin/SXS make an effect on the average size of the corresponding vesicles in the oil-water emulsion. In addition, egg lecithin considered as hydrophobic substance, sodium xylenesulfonate (SXS) acts as the hydrophilic matter. As the ratio goes up, the average vesicle size increases correspondingly.

Stability can be improved by offering suitable stabilizer because the surfactants act as film former and barrier to the release at internal interface(Khoee and Yaghoobian, 2008). Two charged biopolymers, whey protein isolate (WPI) and enzymatic modified pectins, interacted in aqueous solution to form a charge-charge complex which acts as a hydrophilic polymeric steric stabilizer improving the multiple emulsion stability .Regulating the conditions to get the result of most relatively stable condition. For example, as pH can determin the size of the complex ,when pH =6, the most stable double emulsion are gained because of the smallest droplet size, the lowest creaming, highest yield, and minimized water transport(Lutz et al., 2009).

Henry et al. (2009) have studied six emulsifiers in their experiments, it is shown that as the amount of emulsifier increased, the phenomenon of coalescence occurs go down. Furthermore, droplet size is dependent on both break-up and re-coalescence events in emulsification, for example, when the surfactant concentration is lower, the droplet size is prone to a result of multiple break-up events. It is shown in the results of experiments that the frequency of droplet coalescence is decreased to a minimum as the process of preparation is under an optimal surfactant concentration, which balances the formation of the smaller possible droplets and relatively stable in preparation and long time storage.

On the basic of experimental results which is analyzed by equilibrium phase diagram as well as observed through polarization microscopy, Yihan Liu et al. (2009) have got the conclusion that certain type of multiple emulsions which a liquid crystal can be formed by the surfactant with water are more stable compared to counterparts with no liquid crystals exist in the surfactant but prepared in the same condition(Liu and Friberg, 2009).

Garti and Aserin (1996) propose that macromolecules together with monomeric surfactants can be served as steric stabilizers to improve the stability of multiple emulsions. The synthetic polymeric surfactants are ideal interfacial barrier to improve thermodynamic stability and entrapment, which is very helpful in reducing release rate of entrapped additives,and preparing smaller double emulsions with long-time stability. Take WPI-polysaccharide conjugates as an example, compared with monomeric surfactants used only, the application of polymeric emulsifiers results in better encapsulation and controlled release of addenda (Benichou et al., 2006).

Transport mechanism in multiple emulsions

Various kinds of possible mechanisms have been illustrated to interpret how the substances transport through the oil phase. Oil soluble substances just transport through the oil phase by diffusion which is served as controlled mechanism. Many factors contribute to the transport rate, such as the properties of oil phase, the nature of ingredients, and the conditions of aqueous phase (Chang et al., 1987) .In the previous study, it is found that water and water soluble substance can easily migrate through the oil phase. Kita et al. (1977) demonstrate that two possible mechanism can be applied to interpret the phenomenon of transportation: (1) reverse micelle transport; (2) diffusion across a very thin lamella.

Cheng et al. (2006) demonstrate that both Cl- and Ag+ can transport through a thick oil film through observing and measuring the formation of AgCl precipitate in the W1/O/W2 multiple emulsion. Ions can not transport through the oil film which is very thin (<1 μm), however, it is interesting that ions can transport through thick oil film, which is clearly observed by using a capillary video microscopy technique. The “reverse micelle transport” mechanism which is shown in Fig.6 can be used to explain this phenomenon. The amphiphilic molecules means one end has the likelihood to aqueous phase while the other end prone to oil phase, reverse micelle means hydrophilic ends assemble together to form the center while the hydrophobic ends stretch into the oil phase. The water soluble ingredients can be transported inside the center of reverse micelle through the oil phase. If the oil film is too thin, the ions can not form reverse micelle, the thickness of oil film is necessary to provide the space of reverse micelle The thickness of oil film has no influence on the rate of migration, moreover, the ions are prone to migrate to from lower salinity to higher salinity in the aqueous phase( Cheng et al., 2006).

According to Garti(1996), when the oil phase is extremely thin, due to the fluctuation of the thickness of oil layer, the water and water soluble substance can transport through the oil layer as the mechanism of diffusion across a very thin lamella. As Fig.7 shown, the oil film is very thin, the ingredients can not transport through it by reverse micelle transport mechanism, so the mechanism is named as getting through lamella which can be considered as simple diffusion through the oil film. Of course, there exist the hydrophobic surfactants in the oil film, the two ends (hydrophobic end and hydrophilic end) can be clearly seen in the Fig.7.

According to Wen et al., (2000), under the experimental conditions used, it is concluded that the water transport rate in W1/O/W2 multiple emulsions is controlled by interfacial processes, rather than diffusion controlled mechanism which is usually used to explain the phenomenon of transport in many research results. When the distance between W1 and W2 is at visual contact, the hydrated surfactant mechanism applied to explain the occurrence of water transport. Whereas the distance between W1and W2 reaches the minimum, reverse micelle is considered as the mechanism of the occurrence of migration.

Encapsulation efficiency and release rate

Encapsulation efficiency is served as one of the aspects to investigate the stability of W1/O/W2 multiple emulsions. In the first step, the active markers which is used for easy measurement together with the active ingredients are entrapped into the W1 phase in W1/O emulsion. After formation of W1/O/W2 multiple emulsions, the percentage of markers which still exist in the W1 phase is defined as encapsulation efficiency (O’Regan and Mulvihill, 2009).As the time of storage or exposure of the W1/O/W2 multiple emulsion goes on , the encapsulation efficiency will be measured, which is considered as the method to investigate release rate.

Khoee and Yaghoobian (2008) propose that regulating quantities and characteristics of surfactants can improve encapsulation efficiency and slow down release rate of entrapped matter (penicillin-G). Many approaches have been put forward to modify the conditions in the process of preparing multiple emulsions, for example, the selection of the suitable oil phase as the viscosity is important to the stability as well as the type and concentration of surfactants which are used in oil and external aqueous phase. W1/O/W 2 emulsion solvent evaporation technique is successful applied to encapsulated Penicillin-G into PBA nanocapsules.

In order to measure the encapsulation efficiency, O’ Regan and Mulvihillby (2009) propose that after separating W1/O/W2 multiple emulsions which will be put in vivaspin cells into inner phase (W1/O) and external aqueous phase (W2) by centrifugation, the concentration of the marker which comes from the internal aqueous phase to the external phase (W2) will be measured by absorbance spectrophotometry. The encapsulation efficiency stands for the concentration of markers still remain in the internal aqueous phase. So, the result can be calculated as

That is to say, the lower concentration of markers measured in W2, the higher encapsulation efficiency.

Though the previous research, it is observed that the active ingredients partly release from the inner phase, two main mechanism can put forward to explain the phenomenon. (1) Diffusion. If the water or water-soluble substances are included in diffusion-controlled release, reverse micellar transport created by

the hydrophobic emulsifier or simple diffusion across the oil phase linked with osmotic differences between internal and external water phases, are considered as the primary two way how the ingredient come to the outer phase, on the basic of several experiments ,many factors can influence the release rate, for example, molecular weight of the ingredient, the type and concentration of surfactant ; (2) membrane rupture, which is connected with the physical stability of the multiple emulsions, that is to say, the ingredients come out from the inner phase by the rupture of oil film.( Vasiljevic et al., 2005; Schmidts et al., 2009) Which mechanism plays a vital role in the process of release depends on the condition of multiple emulsions, it is possible to shift from one type to the other type by modifying the behaviour of multiple emulsions, such as the nature of active ingredients ,the properties of emulsifiers and so on( Pays et al., 2001).

M. Bonnet et al. (2007) illustrate that they choose magnesium as a marker which is contained in the internal aqueous phase to investigate the release rate of entrapped ingredients from W1/O/W2 multiple emulsions.

Pays et al .(2001) point out that studying the two mechanism in the process of release , they are all exist in the transportation of ingredients from the internal aqueous phase to the external aqueous, on the basic of experimental results from Pays et al. (2001), the coalescence of thin liquid film and globule surfaces plays a vital role in certain multiple emulsions because this mechanism is the rate-determining one which holds a high frequency to occur. As Bjerregaard et al. (1999) demonstrate that the release of glucose was dominated by diffusion through the oil phase rather than membrane rupture.Two coentrapped markers, glucose and inulin ,are used to determine the release kinetics in the experiments. The concentration of glucose is ranged from 40 to 280 mM and the concentration of inulin is remained at 0.4 mM . The release rates of glucose and inulin are almost linear, The result of permeability coefficients for glucose are much larger than for inulin because of the difference of their molecule weight.

Phase ratio of w/o emulsion is regarded as one of the factors which contribute to the release rate, as the increasing of water content, the amounts of droplets per unit of emulsions go up according to , which result in the increase of release rate. In above equation, N stands for the number of droplets, means the volume ratio of W/O emulsion, r is the radius of droplet( Bjerregaard et al., 1999).

In addition, osmotic behavior is another parameter to study the release rate as the oil film are commonly regarded as a semipermeable membrane. There exist an osmotic gradient in this semipermeable membrane, which determine the direction of the flux of water. Matsumoto and Kohda (1980) point out the equation below to explain the relationship in osmotic behavior in W1/O/W2 multiple emulsions:

Jw is the flux of water, Lp means hydrodynamic coefficient of the oily membrane, A stands for the area of the membrane, T is absolute temperature, g1 is osmotic coefficients of electrolyte solutions of concentrations c1, V is partial

molar volume