Spontaneous droplet formation techniques for monodisperse emulsions preparation – Perspectives for food applications

Journal of Food Engineering 107 (2011) 334–346

Contents lists available at ScienceDirect

Journal of Food Engineering journal homepage: www.elsevier.com/locate/jfoodeng

Spontaneous droplet formation techniques for monodisperse emulsions preparation – Perspectives for food applications Abid Aslam Maan ⇑, Karin Schroën, Remko Boom Wageningen University, Food Process Engineering Group, Bomenweg 2, 6703 HD Wageningen, The Netherlands

a r t i c l e

i n f o

Article history: Received 3 March 2011 Received in revised form 2 July 2011 Accepted 7 July 2011 Available online 23 July 2011 Keywords: Emulsification Microchannels EDGE Monodisperse emulsions

a b s t r a c t Spontaneous droplet formation through Laplace pressure differences is a simple method for making monodisperse emulsions and is claimed to be suited for shear and temperature sensitive products, and those requiring high monodispersity. Techniques belonging to this category include (grooved) microchannel emulsification, straight-through microchannel emulsification, and EDGE (Edge-based Droplet GEneration). In this paper, an overview is given of the process, and design parameters that play a role in microchannel emulsification including their effect on droplet size and distribution. Besides, various products made by microchannel emulsification are discussed. Industrial microchannel emulsification is still not possible due to the low production rates. The new EDGE mechanism seems an interesting development, since it promises larger throughputs per droplet formation unit, better scalability, and shows robust operation with practical, food-grade components. However, for spontaneous emulsification techniques to be used on large scale, improvements in construction materials (including surface modification) are expected to be of essence. Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction Emulsions play an important role not only in foods, but also in cosmetics pharmaceuticals and petrochemicals. Characteristics of emulsions including rheology, appearance, chemical reactivity, and physical stability are all affected by the size of the droplets and distribution of droplet sizes (Sugiura et al., 2002c,d); coefficient of variation (CV) is mostly used as an index to characterize the size distribution. An emulsion having CV less than 25% is considered as a monodispersed emulsion while the one having CV above this value is considered as a polydispersed emulsion (Romoscanu et al., 2010). Monodispersed emulsions have several applications in science and industry (van Dijke et al., 2010d). Traditional equipments used for emulsions preparation include high pressure homogenizers, ultraAbbreviations: CV, coefficient of variation (%); DPL, Laplace pressure (Pa); cow, interfacial tension (N m1); R1R2, radii of curvature (m); h, contact angle (°); DPBT, breakthrough pressure (Pa); g, viscosity (Pa s); n, viscosity ratio (dispersed phase/ continuous phase); GMC, grooved microchannels; STMC, straight-through microchannels; Lch, length of the channel (m); Wch, width of channel (m); Dch, depth of channel (m); LT, length of terrace (m); WT, width of terrace (m); Lst, longer length of straight-through channel (symmetric)/microslot (asymmetric) (m); Sst, shorter length of straight-through channel (symmetric)/microslot (asymmetric) (m); Dst, depth of straight-through channel (symmetric)/microslot (asymmetric) (m); D, diameter of cylindrical channel (asymmetric) (m); Lp, length of plateau (m); Wp, width of plateau (m); Hp, height of plateau (m); Np, total number of plateaus. ⇑ Corresponding author. Tel.: +31 317 483770; fax: +31 317 482237. E-mail address: [email protected] (A.A. Maan). 0260-8774/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.jfoodeng.2011.07.008

sound homogenizers and rotor–stator systems (stirred vessels, colloid mills or toothed disk dispersing machines) (see Fig. 1) (Sugiura et al., 2002b). These instruments apply a high shear stress to deform and disrupt the large droplets into smaller ones. Only 1–5% of the applied energy is used for droplet formation and remaining is lost as heat (van Dijke et al., 2010c) which can cause temperature and shear sensitive ingredients (starch, proteins, etc.) to lose their functional properties (Charcosset et al., 2004). In addition, prepared emulsions are polydispersed with coefficient of variation (CV) of around 40% (Saito et al., 2006), which makes these emulsions intrinsically unstable. In the last two decades, alternative emulsification techniques have been proposed, which can provide better control over droplet size while consuming less energy. Nakashima et al. (1991) proposed the cross-flow membrane emulsification technique to produce monodisperse emulsions at much lower mechanical stress which results in low energy requirements. A schematic diagram of the process of cross-flow membrane emulsification is shown in Fig. 2. The dispersed phase is pressed through the pores of a membrane and droplets are formed at the pore openings, where the droplets are sheared off and carried away by the cross-flowing continuous phase. This technique is useful for producing monodisperse emulsions with a coefficient of variation of about 10% (Sugiura et al., 2001b), but it is feasible only for dilute emulsions (<10%) that can be produced in one step; recirculation that may be used to increase the dispersed phase fraction, induces wider droplet size distributions.

335

A.A. Maan et al. / Journal of Food Engineering 107 (2011) 334–346

Fig. 1. Schematic illustration of traditional emulsification equipments. High pressure homogenizer (a) ultrasound homogenizer (b) and rotor stator system (c).

Pressurized dispersed phase Membrane

Continuous phase

Emulsion

Membrane Pressurized dispersed phase Fig. 2. Schematic illustration of cross flow membrane emulsification.

Within the field of microtechnology, new emulsification techniques have emerged. One class of devices is similar to cross-flow membrane emulsification and uses shear forces to form droplets. Some examples are T-, Y- shaped microchannel junctions (see Fig. 3a and b respectively), in which a cross-flow is used, and flow-focusing devices (Fig. 3c and d) in which co-flow of phases exerts extensional shear. Another class of devices uses so-called spontaneous droplet formation, of which the most studied form is microchannel emulsification (Fig. 3e and f). Spontaneous emulsification is an interesting technique to develop further because it renders emulsions that can be high in dispersed phase fraction while still having a sharp droplet size distribution (CV  5%) (Sugiura et al., 2002d); the processing time is proportional to the dispersed phase fraction. Microchannel emulsification is a relatively new technique for preparation of monodisperse emulsions. Both, terrace-shaped structures (grooved) (Sugiura et al., 2001b) and straight-through microchannel arrays (Kobayashi et al., 2002a,b) (see Fig. 3e and f respectively) have been extensively reported in literature. As mentioned, the distinguishing feature of this technique is that no shear forces are needed to form droplets, and further, the size of droplets is mostly determined by the microchannel geometry (see also Fig. 4), and to a lesser extent by the to-be-disperse-phase flow. Interfacial tension is used as a driving force for droplet formation; and less energy (a factor of 10–100 less) as compared to the conventional techniques is needed (Sugiura et al., 2001b), for the production of various products such as O/W emulsions, W/O emulsions, lipid microparticles, polymer microparticles or microcapsules (Sugiura et al., 2004a). This paper aims to provide an overview of microchannel emulsification including, process principles, process parameters, applications, and an outlook on new emulsification technology based on spontaneous droplet formation that can be used in the production of food (related) products.

2. Microchannel emulsification A (grooved) microchannel consists of a narrow channel fabricated in a microchip (made of silicon or a polymer) covered tightly with a (glass) plate. The channel ends into a slit like structure, the terrace, which leads to a deep continuous phase/emulsion channel. A schematic diagram of the process of droplet formation through a typical microchannel design is shown in Fig. 4. On applying enough pressure, the dispersed phase flows through the narrow channel and spreads over the terrace in the form of a flattened disk-like shape (Fig. 4a–c). Curvature of the interface produces a pressure difference between the two phases called Laplace pressure which is defined as (van Dijke et al., 2010d).

DPL ¼ cow 



 1 1 cos h þ R1 R2

ð1Þ

Where DPL is the Laplace pressure (Pa), cow is the interfacial tension between the oil and water phases (N m1), R1 and R2 are the two radii of curvature of the interface (m) and h is the contact angle (). On reaching the end of the terrace (Fig. 4d) the tip of the dispersed phase leaps over into the continuous phase channel, and assumes a spherical shape which is energetically favorable (Fig. 4e), and detaches spontaneously (Fig. 4f) when the droplet has become large enough. (Laplace) pressure differences are the driving force behind microchannel emulsification (Sugiura et al., 2001b), as is indicated by the curvatures in Fig. 4g–i, and as could also be derived from CFD simulations reported by Kobayashi et al. (2004a) and van Dijke et al. (2008). In the later publication, design rules are shown that are based on an analytical model that uses local (Laplace) pressure differences (see Section 3). A straight-through microchannel device (Kobayashi et al., 2002a,b) consists of a large number of through holes fabricated

336

A.A. Maan et al. / Journal of Food Engineering 107 (2011) 334–346

Fig. 3. Some examples of microfluidic devices which are capable of producing monodisperse droplets. (a) T-Junctions (b) Y-Junctions (c) Flow–focusing microcapillary device (Reprinted with permission from Utada et al. (2007)) (d) Coflow system (droplet growth (a) and separation (b)) (Reprinted from Umbanhowar et al. (2000)) (e) Grooved microchannel (Reprinted from van Dijke et al. (2008)) (f) Straight-through microchannel device (Reprinted from Kobayashi et al. (2005b)).

on a single microchip (Fig. 3f), that works according to the same mechanism described for grooved microchannels. The dispersed phase is pressurized from one side of the device, pressed through the holes, and forms droplets on other side. The channels are constructed such that the continuous phase can intrude into the holes, to help in the formation and shrinkage of the neck inside the channel, and ultimately, formation of a droplet (Kobayashi et al., 2004a).

applied pressure will be discussed, followed by liquid (viscosity) and ingredient properties (surfactants, etc.) that will be discussed in the light of the emulsification process, and finally surface properties will be touched upon. 3.1. Geometry

3. Process parameters

3.1.1. Grooved microchannels The grooved microchannels consist of a channel and a terrace part (see Fig. 5a).

As mentioned previously, two distinct geometries are used for microchannel emulsification, grooved systems, and straightthrough systems. Here we first describe the geometry of both systems together with the typical dimensions that are related to both. After that, the properties related to the process operation such as

3.1.1.1. Terrace. Terrace geometry is defined with its length (LT), width (WT) and height (HT). The terrace causes formation of a neck and promotes droplet formation. Kobayashi et al. (2001) showed that an array of microchannels with terraces gave monodisperse microspheres, while without terrace polydispersed microspheres

A.A. Maan et al. / Journal of Food Engineering 107 (2011) 334–346

337

Fig. 4. Droplet formation process in microchannel emulsification (a–f), together with cross sections in various positions in the microchannel (g–i) (Reprinted from Sugiura et al. (2001b)).

were obtained. In general, droplet size increases with increasing available volume on the terrace without affecting the monodispersity of droplets (Kawakatsu et al., 2000; Sugiura et al., 2002e); basically, the entire terrace empties to form the droplet.

3.1.1.2. Channel. The channel geometry is defined with its length (Lch), width (Wch) and depth (Dch), and especially the length influences the pressure range within which monodisperse emulsions can be obtained, but not the droplet size (Liu et al., 2005a; Sugiura et al., 2002b). This was also concluded from the analytical model presented by van Dijke et al. (2008); the pressure needed for flow through a longer channel is higher, while the Laplace pressure on the terrace and in the droplet will be unaffected, and this leads to higher monodispersity for longer channels because the system is less pressure sensitive. Sugiura et al. (2002b) reported that the droplet size is not affected by channel width, but Kawakatsu et al. (2000) and Kobay-

ashi et al. (2001) found an increase in droplet diameter by increasing channel width. Similar results have been reported by Sugiura et al. (2002c). In general, an increase in the equivalent hydrodynamic diameter of the channel leads to a narrower pressure range for stable droplet formation (see also flow velocity section) and monodispersity decreases as a result thereof. This is also true for other variations in channel dimensions as studied by Liu et al. (2005a) for convexes and Kawakatsu et al. (2000) for triangular channels.

3.1.2. Straight-through microchannels The geometry of straight-through channels is identified as shorter length (S), longer length (L) and the depth (D) as shown in Fig. 5b. Two distinct designs have been reported in literature, one with equal dimension all through the structure (symmetric), and one in which a narrow channel ends into an area with these dimension (asymmetric, and similar to the terrace system) (see

Channel

LT Lch

Wch

WT Dch HT Terrace

Microchip

(a)

(b)

Fig. 5. Schematic illustration of microchannel geometries. (a) Grooved microchannels (b) Straight-through microchannels.

338

A.A. Maan et al. / Journal of Food Engineering 107 (2011) 334–346

Fig. 6) (Vladisavljevic et al., 2008). Monodispersed droplets are produced from channels with a slot aspect ratio (length/width ratio) above 3–3.5. At lower aspect ratio, the cross section of the channel is completely occupied by the growing droplet, which prevents the inflow of the continuous phase, and therewith, droplet formation (Kobayashi et al., 2004a; Vladisavljevic´ et al., 2010), again indicating the importance of the ‘terrace’ that induces instability in the to-be-dispersed phase. Monodispersed emulsions were reported for asymmetric and oblong symmetric channels; only if the resistance of the supply channel became too high, polydisperse emulsions were observed and even continuous outflow. 3.2. Pressure and dispersed phase flow rate Microchannel emulsification requires a pressure to be exerted on the dispersed phase to cause it to flow through the channel, and subsequently on the terrace. When the applied pressure reaches a certain value, (which is very low compared to pressures applied in e.g. high pressure homogenizers); the droplets begin to form from the terrace end. This pressure at which the droplet formation starts is called breakthrough pressure and can be estimated with (Sugiura et al., 2002d):

DPBT ¼

4cow cosðhÞ H

ð2Þ

where, DPBT is the breakthrough pressure (Pa), cow is the interfacial tension (N m1), and h is the contact angle of the interface with the channel surface () and H is the terrace height (m: smallest dimension of the terrace). Upon increasing the pressure (liquid flow velocity), the droplet frequency will increase and the size of the droplets (and distribution) will also increase slightly as shown in Fig. 7. Please note that flow velocity data need to be interpreted with care, because the applied pump rate that is normally used on the x-axis may be quite different from the actual flow rate due to pressure build-up inside the microchips.

Fig. 6. schematic representation of symmetric (a and b) and asymmetric microchannels (c) (Reprinted from Vladisavljevic et al. (2008)).

With increasing pressure, up to a certain critical velocity (encircled points in Fig. 7b) the droplets stay monodisperse, but above this critical pressure/flow velocity, the droplet diameter increases more rapidly (as can be seen in the Figure) and monodispersity decreases, and eventually blow-up occurs (i.e. the phase flows out continuously). At these pressures, the flux from the supply channel, across the terrace to the droplet is so large that it does not allow the neck, that keeps the droplet connected to the terrace, to collapse, as described by the flux criterion defined by van Dijke et al. (2008), as summarized in Appendix B. The effect of the applied pressure (and flow velocity) on droplet size depends on various parameters such as microchannel geometry (van Dijke et al., 2008), viscosity ratio of dispersed and continuous phase (van Dijke et al., 2010b) and the type and concentration of surfactants (Sugiura et al., 2004a; Sugiura et al., 2002b). More details are given in the respective sections. 3.3. Liquid and ingredient properties 3.3.1. Viscosity The viscosities of dispersed and continuous phases have important effect on the performance of microchannel emulsification processes. The average droplet diameter of W/O emulsions was reported to increase with increasing continuous phase viscosity (Kawakatsu et al., 2001b; Liu et al., 2004) and for O/W emulsions, the average droplet diameter was reported to decrease with increasing dispersed phase viscosity (Kawakatsu et al., 2001b; Vladisavljevic et al., 2008). All these effects may be covered by the observations by van Dijke et al. (2010b), who linked the viscosity ratio (viscosity of dispersed phase/viscosity of continuous phase) to the droplet size. In general, the droplet size scales with the viscosity ratio; at high viscosity ratio, the droplet size is constant, and at low viscosity ratio, the droplet size increases (see Fig. 8). This can be interpreted as follows: as long as the continuous phase can flow onto the terrace freely (compared to the to-be-dispersed phase flow), the neck will break rapidly, and the droplet size will not be influenced. If the continuous phase is very viscous, this will prevent the neck from collapsing rapidly; keep the droplet connected to the feed channel for a longer time, and lead to larger droplets. Temperature has a direct affect on the viscosities of dispersed and continuous phases but the viscosity ratio will not be greatly affected. Decrease in viscosity of phases increased the droplet formation frequency, but as long as the values for viscosity ratio (gd/gc) remained above the critical viscosity ratio (and the interfacial tension is not too much influenced), the droplet size remains unaffected. Hence, the temperature can be used as a tool to tune the droplet diameters and droplet productivity. On the other hand, it makes the emulsification process more complex since it also influences e.g. interfacial tension (see next Section 3.3.2); and temperature sensitive components should not be affected. 3.3.2. Surfactants and interfacial tension Surfactants play two important roles in the emulsification process; firstly, they lower the interfacial tension and stabilize the droplet after formation, i.e. prevent coalescence and/or aggregation of emulsions. As a side effect, surfactants may influence the wettability of the emulsification device, therewith indirectly influencing local pressures, and through this, the droplet size. While in classic emulsification methods a low interfacial tension facilitates droplet formation, in microchannel emulsification, this is not the case. At high interfacial tension, the pressure differences between the feed channel and droplet are higher, and already at small size, the flux criterion of van Dijke et al. (2008) will be met. At low interfacial tension, the droplet needs to become bigger to result in a similar pressure difference. In general, it is expected that for microchan-

339

A.A. Maan et al. / Journal of Food Engineering 107 (2011) 334–346

20

80

15

60 10 40

CV (%)

Average Diameter (µm)

(a) 100

5

20

0

0 0

0.5

1

1.5

2

2.5

3

Pressure (KPa)

Average diameter (µm)

(b)100 90 80 70 60 50 40 30 20 10 0 0

1

2

3

4

Flow velocity (mm/s) Fig. 7. Effect of pressure and flow velocity on droplet formation: (a) Average droplet diameters (N, j) and coefficient of variation (CV) (D, h) of water/decane emulsions, containing surfactants CR310 (N, D) and PO500 (j, h), as a function of applied pressure (data taken from Liu et al. (2004)) (b) Average droplet diameters as a function of flow velocity of dispersed phase using different microchannels (MC-2 () Dch = 2 lm, Wch = 3.3 lm, Lch = 7.7 lm, LT = 15 lm; MC-4 (N) Dch = 4 lm, Wch = 4.7 lm, Lch = 14 lm, LT = 28 lm; MC-8 (j) Dch = 8 lm, Wch = 8.3 lm, Lch = 32 lm, LT = 57 lm; MC-16 (j) Dch = 16 lm, Wch = 16 lm, Lch = 68 lm, LT = 113 lm) (data taken from Sugiura et al. (2002a)).

Dimensionless droplet diameter D [-]

6 5 4 3 2

hexadecane Silicon oil 200 Silicon oil 1000

1 0 1.00E-01

1.00E+00

1.00E+01

1.00E+02

Soybean oil Silicon oil 500 Silicon oil 5000

1.00E+03

1.00E+04

Fig. 8. Effect of viscosity ratio (gd/gc) on dimensionless droplet diameter D (resultant droplet diameter Ddrop/height of the terrace H) (data taken from van Dijke et al. (2010b)).

nels the droplet formation process is thus slow that the surface will be saturated, i.e. at equilibrium interfacial tension (van Dijke et al., 2010c). Whether the interfacial tension is lowered during the time of droplet formation, depends on the type of surfactant and its concentration (Joscelyne et al., 2000). Vladisavljevic et al. (2008) studied the effect of SDS (sodium dodecyl sulfate) concentration, on the preparation of oil-in-water emulsions using soybean oil as dispersed phase, with straight-through microchannels. Monodisperse droplets were formed in the range of 0.2–0.5 wt%; at lower concentration, the droplets coalesced, while at higher concentrations, satellite droplets were formed, therewith reducing monodispersity. Satellite droplets are generated as a result of imbalance of capillary forces during break-off of primary droplets (Tan et al., 2006) and has been reported in many droplet generation devices (Anna et al., 2003; Tan et al., 2004).

Kobayashi et al. (2003) investigated the effect of differently charged emulsifiers on the preparation of oil-in-water emulsions with straight-through microchannels. Anionic surfactant SDS, nonionic surfactant Tween 20 (polyoxyethylene (20) sorbitan monolaurate) and a cationic surfactant CTAB (cetyltrimethylammonium bromide) were dissolved in continuous phase while a cationic surfactant TOMAC (Tri-n-octylmethylammonium chloride) was dissolved in dispersed phase (soybean oil). Monodisperse emulsions were successfully produced by using anionic and nonionic surfactants, while cationic surfactants resulted in polydisperse emulsions and complete wetting of the channel surface by the dispersed phase, indicating that surface properties need to be considered in combination with surfactant properties. Several investigations with food grade emulsifiers have been reported. Saito et al. (2005) made oil-in-water emulsions using bovine serum albumin (BSA), b-lactoglobulin, c-globulin, lysozyme,

340

A.A. Maan et al. / Journal of Food Engineering 107 (2011) 334–346

soybean flour, whey protein, and egg white protein. Stable monodispersed emulsions could only be prepared using BSA, b-lactoglobulin, soybean flour and whey protein (see also Section 3.4). Besides, the ingredients of the emulsion influence derived parameters such as the surface tension and emulsion stability. Fujiu et al. (2010) recently investigated the effect of temperature (10–70 °C) on microchannel emulsification. At higher temperature, the breakthrough pressure (DPBT) increased because of higher interfacial tension (see Eq. (2)), and smaller droplets were formed, albeit at higher frequency due to the lower viscosity of the to-bedispersed phase. 3.4. Surface properties For stable production of monodisperse emulsion droplets through microchannels, the continuous phase should wet the channel surface; hence the surface should be hydrophilic for O/W emulsification and hydrophobic for W/O emulsification (Kobayashi et al., 2008b). That also explains why Liu et al. (2004) using a hydrophobic acrylic microchannel for the preparation of O/W emulsion found continuous out-flow of oil without any droplet formation. For the production of microchannels, mostly silicon-based materials are used, and various researchers investigated chemical modification of the surfaces. For example, silicon is hydrophobic but can be made hydrophilic by depositing a silicon-oxide layer through plasma treatment (Kobayashi et al., 2008b). Silicon can also be made more hydrophobic through silane coupler reagents, such as octadecyltrichlorosilane, and subsequent heating at 110 °C for 1 h (Liu et al., 2004). However, silanization is not a permanent surface modification method, and may not be used for food and pharmaceutical applications (Kobayashi et al., 2008a). For permanent modification, the covalent methods developed by Arafat et al. (2004), Arafat et al. (2007) and Rosso et al. (2009) seem more appropriate, although these methods have only been applied on flat surfaces until now. Also acrylic microchannels, which are originally hydrophobic and suitable for W/O emulsification, can be modified in order to make them suitable for O/W emulsification by a procedure described by Liu et al. (2005b). Monodisperse emulsions were produced when using an Exceval coating and SiO2 vacuum deposition while Lipidure-PMB coating produced polydisperse emulsions. Clearly, surface properties need to be chosen carefully, and the durability of the modified layer needs to be evaluated carefully.

As mentioned in the previous Section 3.3.2, the surface of the microchannels may also be affected by the surfactants. Various examples are summarized in Table 1, and they clearly illustrate the dual effect of surfactants which are needed to stabilize the oil/water interface, but affect the (wettability of the) microchannel surface mostly negatively, but also sometime positively. Saito et al. (2005) could not prepare oil-in-water emulsions with c-globulin, lysozyme and egg white proteins because they influenced surface wettability, resulting in high water contact angles that prevent water intrusion on the terrace. In some other investigations, polydisperse O/W emulsions with large droplet sizes were obtained when surfactants adsorbed to the channel surface rendering it non-uniform wettability or even making it hydrophobic (Kobayashi et al., 2003; Sugiura et al., 2000). Tong et al. (2002) investigated the effect of soybean and egg yolk lecithin (dissolved in the oil phase), and found that the oil droplets coalesced and the microchannel surface lost its hydrophilicity due to lecithin adsorption, resulting in the continuous outflow of the oil phase. Similar behavior for lecithin has been reported by Sugiura et al. (2000). For the preparation of lipid microspheres and polymeric microspheres, the same authors (Sugiura et al., 2001a) found that the microstructure was not wetted uniformly below 0.2 wt% SDS; in this case, a certain amount of surfactant is needed to maintain appropriate surface wetting properties. Changes in wettability will be especially important in the production of foods that contain many different surface active components. Which combination works best is hard to predict beforehand, given the multitude of interactions that play a role. E.g. the pH can influence the properties of the surface and those of the components, as reported by Huisman et al. (1998) and Nakagawa et al. (2004). Huisman et al. (1998) used different ceramic membranes, and the charge of the membranes determines which surfactants can adsorb. Nakagawa et al. (2004) investigated the influence of pH on the preparation of gelatin/acacia complex coacervate microcapsules, and regular sized microcapsules could not be obtained at pH 4.0, because gelatin molecules (isoelectric point 5.1) were positively charged and interacted with the negatively charged surface of microchannel, rendering it hydrophobic. At a pH above the isoelectric point of gelatin, it was possible to produce monodisperse droplets due to the negative charge of the gelatin. Saito et al. (2005) observed similar behavior for BSA at low pH (3 and 4), where the microchannel surface was wetted by the dispersed phase; clearly not only basic adsorption is important but also charge interactions (mostly with the microchannel) are crucial

Table 1 Effect of type and concentration of surfactants on microchannel emulsification. Type of emulsion

Oil-in-water

Oil-in-water

Oil-in-water

Surfactant type and concentration

a b

References

Coalescence Satellite droplet formation Monodispersed emulsion Wetting of dispersed phase on channel surface

Vladisavljevic et al. (2008)

Type

Conc. (Wt%)

SDS (Wc)a SDS (Wc) SDS (Wc) c-globulin (Wc) Lysozyme (Wc) Egg white (Wc) BSA (Wc) CTAB (Wc)

Anionic Anionic Anionic – – – – Cationic

0.01–0.02 >0.5 0.2–0.5 0.5 0.5 0.5 0.5 1.0

TOMAC (Od) Tween 20 (Wc) SDS (Wc)

Cationic Nonionic Anionic

2.0 <0.2

Monodispersed emulsion Wetting of dispersed phase on channel surface and satellite droplet formation Wetting of dispersed phase on channel surface Monodisperse emulsion Wetting of dispersed phase on channel surface

Soybean and egg yolk lecithin

–

0.3

Wetting of dispersed phase on channel surface and coalescence

b

Polymeric microspheres Oil-in-water microspheres

Emulsification behavior

Surfactant

Dissolved in continuous water phase. Dissolved in dispersed oil phase.

Saito et al. (2005)

Kobayashi et al. (2003)

Sugiura et al. (2001a) Tong et al. (2002)

A.A. Maan et al. / Journal of Food Engineering 107 (2011) 334–346

for maintaining emulsification. For this, surface modification methods, as suggested by Rosso et al. (2009) which prevent protein adsorption could be of great relevance. For industrial applications, the use of polymer microchannels has an advantage over silicon in the sense that it is stronger and tougher than silicon. In addition, polymer is less expensive and can be processed more easily (Liu et al., 2004, 2005a). However, the material from which the microstructures are made also has to satisfy other needs (appropriate surface properties, and it should allow preparation of small channels). Materials with high mechanical stability, especially metals which are expected to be of great industrial relevance, still need to be investigated. This is probably because the precise fabrication needed for microstructures is currently not possible. However, given the opportunities that arise from emulsification with microchannels, it can be expected that metal fabrication will grow into a viable concept in near future.

4. Products produced by microchannel emulsification Microchannel emulsification technique has been successfully applied for the preparation of simple emulsions, multiple emulsions, microspheres, and microcapsules. Here we shortly discuss these products and where possible we compare with similar products obtained by more traditional emulsification technology. Oil in water emulsions were reported by many different authors, while water-in-oil emulsions are more sparingly reported e.g. (Kawakatsu et al., 2001b; Liu et al., 2004; Liu et al., 2005a). A complete overview of the data available in literature at the time of writing can be found in Table A1 for grooved microchannel systems, and in Table A2 for straight through systems. Saito et al. (2006) found that O/W emulsions prepared by microchannel technology showed higher stability than those obtained by homogenization, which was attributed to the monodispersity of the emulsions. A multiple emulsion is an emulsion in an emulsion (van der Graaf et al., 2005), and they cannot or hardly be prepared by classic emulsification methods, because the second emulsification step is prone to destroy the primary emulsion. Literature is available on the preparation of W/O/W emulsions (Kawakatsu et al., 2001a; Kobayashi et al., 2005a; Sugiura et al., 2004b) but no literature has been reported until now on the preparation of O/W/O emulsions by microchannel emulsification probably because they are less interesting from an application point of view. Typically, microchannel emulsification is used as a second step for dispersion of the primary emulsion into the continuous phase. Monodisperse

341

emulsions were successfully prepared with little or no leakage of internal water phase (Kobayashi et al., 2005a), which makes this technology a very interesting preparation method for such products, which may be used as low-calorie products or as controlled release vehicles. Microspheres are solid particles which can be utilized in food, cosmetics and pharmaceuticals (Bor and Calvin, 1997; Friberg and Larsson, 1997); microchannel emulsification was reported to successfully prepare monodisperse microspheres of sizes ranging from several to 100 lm (Sugiura et al., 2001a). Sugiura et al. (2002e) prepared divinylbenzene microspheres by combining microchannel emulsification of the monomer with polymerization, and the resultant microspheres had a narrow size distribution similar to those of polymeric microspheres prepared by seed polymerization. Other examples of microspheres prepared through microchannel emulsification can be found in Tong et al. (2000, 2002) and Sugiura et al. (2000). Microcapsules are hollow microspheres that consist of a polymer wall or coat that covers a core that may contain an active ingredient (Arshady, 1993; Forssell et al., 2006; Vilstrup, 2001), being e.g. a food additive, a biocide, or an adhesive (Charcosset, 2009). Owing to its low energy input, microchannel emulsification can be used for shear sensitive substances like peptides and proteins. Some examples: Gelatin capsules with narrow size distribution were prepared by Iwamoto et al. (2002) using iso-octane containing TGCR (Tetraglycerin condensed ricinoleic acid ester) as a continuous phase. Neves et al. (2008) encapsulated b-carotene dissolved in soybean oil in sugar ester or gelatin, and obtained micrometer sized monodisperse loaded capsules, which were physically stable after 4 months of storage at 5 °C. Simple emulsions can be used for capsules built by layer-by-layer adsorption as reported by Sagis et al. (2008). When starting from monodisperse droplets obtained from microchannels, exact dosage in these capsules is within reach (please note, that this was not the case in (Sagis et al., 2008)).

5. Recent developments and comparison of emulsification techniques The microchannel emulsification technique is claimed to be suitable for the production of ‘supermonodisperse’ emulsions (Sugiura et al., 2001b), however the low throughput of the disperse phase (less than 0.1 ml/h) from grooved microchannels (Vladisavljevic et al., 2008) limit the range of its practical applications. Development of straight-through devices was an effort to increase the

Fig. 9. (a) Droplet formation through a typical EDGE device (Reprinted from van Dijke et al. (2010a)) (b) A typical parallelized EDGE system (reprinted from van Dijke et al. (2009)).

342

A.A. Maan et al. / Journal of Food Engineering 107 (2011) 334–346

Fig. 10. Energy efficiencies of various emulsification systems: () grooved microchannels (GMC) (Sugiura et al., 2000), (e) straight-through microchannels (STMC) (Kobayashi et al., 2002a,b), () EDGE systems (van Dijke et al., 2009), (+) Y-junctions (Steegmans, 2009), (N) membrane emulsification, (s) flat valve homogenizer, (h) orifice valve and (d) microfluidizer (Lambrich et al., 2005).

productivity of the system and monodisperse emulsions at a rate of 65 L/(m2 h) were successfully produced with average droplet diameter of 30 lm (Kobayashi et al., 2002b). However, for narrower microchannels that produce, i.e. droplets below 10 lm which is the size range required in most food applications, the maximum oil flow was only 0.708 L/(m2 h) (Kobayashi et al., 2008b) because of a low percentage of active channels. In some cases, this percentage was below 1%, and upon slight increase of the pressure, the system became unstable. Most probably this is caused by pressure gradients just below the straight-through plate as was reported by Abrahamse et al. (2002) for microsieve systems.

For small droplets (<10 lm) the new EDGE (Edge-based Droplet GEneration) mechanism (van Dijke et al., 2010d) which results in multiple simultaneously formed monodispersed droplets from a single droplet formation unit, can be of great interest. Droplet formation in an EDGE device is shown in Fig. 9a; the dispersed phase is pressurized through the oil channel which spreads over a large flattened area called plateau and on reaching the edge of the plateau, it spontaneously forms monodispersed droplets at several locations. In microchannel emulsification, the terrace would almost completely empty into one droplet, but that is obviously not the case for EDGE where the droplet volume is only a small

350000

Area Required (m 2)

300000 250000 200000 150000 100000 50000 0 EDGE

STMC

GMC

GMC-A

Emulsification system Fig. 11. Area required of different spontaneous emulsification systems (EDGE (van Dijke et al., 2010c, 2009), STMC (Kobayashi et al., 2008b), GMC (Kobayashi et al., 2001) and GMC-A (Kobayashi et al., 2010)) to obtain dispersed phase flux of 1m3/h.).

343

A.A. Maan et al. / Journal of Food Engineering 107 (2011) 334–346

fraction of the total volume of oil available on the plateau, and droplet formation takes place along the entire length of the edge. The droplet size produced by EDGE is determined by the height of the plateau and a scaling factor of 6–8 has been observed. The EDGE technique has only very recently been introduced (2009– 2010); therefore, not a lot of information is available; Table A3 summarizes the investigations made with EDGE systems, and it is clear that the technique gives monodisperse emulsions with coefficients of variation generally around 5–6%. It is worth mentioning that initial studies have shown that the system is robust, more stable under pressure changes than the microchannels are, and it has been operated continuously for 3 consecutive days without the droplet size and monodispersity being affected. Besides it can be parallelized (van Dijke et al., 2010c, 2009) (see Fig. 9b) with all the plateaus completely filled with oil phase at breakthrough pressure and regular sized droplet formation from all plateaus as soon as the oil reaches the edge of the plateaus. Emulsification efficiency of emulsification systems can be compared on energy density which is defined as energy input per unit volume of emulsion. Fig. 10 compares spontaneous emulsification techniques (microchannels and EDGE) with shear based microsystems (membranes and Y-junctions), and traditional emulsification systems (homogenizers and microfluidizers). The energy required for EDGE emulsification is comparable to the grooved and straight-through microchannels, and membrane emulsification, but seems less than needed for shear based and classic emulsification systems, although the picture is obscured because the droplet sizes are not the same and also the droplet size increases by increasing pressure in spontaneous emulsification (also see

Section 3.2). The pressure needed for spontaneous droplet formation is very low (mbar range) compared to that used in classic emulsification devices, and it can be expected that the supplied energy is mostly used for the formation of surfaces, and not dissipated as heat as is the case in homogenizers. Away from monodispersity and energy density, also the microchip area (or volume) required to produce a unit volume of emulsion is an important parameter to evaluate scalability of a technique. Fig. 11 compares the area required of different spontaneous emulsification systems to obtain dispersed phase flux of 1 m3/h. The values are calculated for systems that produce droplets of sizes <10 lm (GMC, STMC and EDGE) (see Tables A1–A3). The area required for EDGE systems is less as compared to the grooved and straight-through microchannels. GMCs require much larger area which is due to the limited number of channels (100– 1500) in most of the investigated systems (Vladisavljevic et al., 2008). Recently, Kobayashi et al. (2010) investigated scale up of grooved microchannels through integration of microchannel arrays on a single microchip (60  60 mm) consisting of 14 arrays and 1.2  104 channels. With this system, they were able to obtain dispersed phase flux of 1.5 mL/h (dav = 10 lm) which is a promising development and can affectively reduce the required area as can be seen in Fig. 11 (GMC-A). Whether any of these approaches can be successfully scaled-up is not clear; however we find that stable operation, and relatively easy fabrication (only one dimension i.e. height of the plateau needs to be precisely defined and maintained) makes EDGE unique among spontaneous emulsification systems and the most likely candidate for scale-up.

Table A1 Microchannel geometry, and resulting droplet diameters under various experimental conditions for grooved microchannels. Microchannel geometry

a b c d e

Dispersed phase (a) Continuous phase (b)

Lcha

Wchb

Dchc

LTd

WTe

lm

lm

lm

lm

lm

–

12

2.0

25

32

– –

3.2 30

1.0 16

5.0 98

7.3 –

– –

30 4.7

16 1.2

240 6.9

– –

199 39.8 120 120 70 –

27.2 10.2 20 ± 0.9 40 ± 1.0 16 ± 0.3 14.4

5.0 1.9 10 20 11 4

– – 30 30 30 29.6

– – – – – –

50

10

5

15

–

150 500 –

10 10 16

5 5 4

15 15 –

– – –

– – – 70

30 30 30 10

16 16 16 2

98 138 240 30

– – – –

250

10

10

50

10

(a) Soybean oil with 3.2 g/L of b-carotene (b) Milli-Q water with 1 wt% sucrose monolaurate (a) Hexadecane (b) Milli-Q with 1% SDS

25.3

–

8

51.7

–

(a) Hexadecane (b) Milli-Q water with 1 wt% SDS

Channel length. Channel width. Channel depth. Terrace length. Terrace width.

(a) Divinyl benzene with 2 wt% benzyl peroxide (b) 0.2 wt% SDS aqueous solution (a) Divinyl benzene with 2 wt% benzyl peroxide (b) 0.2 % SDS aqueous solution (a) Soybean oil with 1.5 wt% Tween 80 (b) physiological saline (a) Refined soybean oil (b) 1.0 wt% Tween 20 in Milli-Q water (a) Water (b) 3 wt% CR310 dissolved in triolein

(a) Triolein (b) 0.3 % SDS aqueous solution (a) Milli-Q water with 3 wt% CR310 (b) Decane/Triolein (20:80 wt/wt)

(a) Air (b) 0.3 wt% SDS aqueous solution (a) Triolein (b) Milli-Q water with 1% SDS

Pressure (KPa)

Droplet diameter (lm)

CV (%)

8.8

10.0

4.4

16.6 1.1

4.2 74.9

9.1 2.8

1.0 5.4

90.2 5.0

2.3 7–9

3.6 7.2 0.8–1.0 0.3–0.6 0.8–1.0 3.5

21.4 9.0 52–60 62–98 53–66 17.8

–

24–37

5.6–7.7

– – –

25–42 27–54 33.6

5.1–6.7 4.6–5.5 1.8

1.16 1.12 1.43 –

64.4 74.6 98.1 9.1

3.4 2.1 2.5 6.2

30

–

34.8

2.5

20 –

2.3 3.1 <8.0 <8.0 <8.0 2.8

References

Sugiura et al. (2001a) Sugiura et al. (2002e) Kobayashi et al. (2001) Kobayashi et al. (2008c) Liu et al. (2004)

(Sugiura et al., 2001b) Liu et al. (2005a)

Yasuno et al. (2004) Sugiura et al. (2002d) Neves et al. (2008) (van Dijke et al., 2008) van Dijke et al. (2010b)

344

A.A. Maan et al. / Journal of Food Engineering 107 (2011) 334–346

6. Conclusion and outlook Spontaneous emulsification with microchannels is a popular method in literature, and the technique is suited for monodisperse emulsions production (CV  5), and for products containing shear and temperature sensitive ingredients (e.g. food). Industrial application is not yet possible, because of scaling issues. The production rates per channel are low, and for the production of small droplets (<10 lm), activation of all channels in straight-through emulsification is a technical challenge.

The new development called EDGE may be the solution to some of the scaling issues related to microchannel emulsification. Its wide plateau allows multiple monodisperse droplets to form from one droplet formation unit, with the size of the droplets only determined by the height of the plateau. Besides, some proof of principle on up-scaling is available, and food ingredients have been successfully applied. Away from the choice of the spontaneous emulsification technique, it is of utmost importance to control the surface properties and maintain appropriate wettability all through emulsification

Table A2 Channel geometry and resulting droplet diameters under various experimental conditions for straight-through microchannels. Channel Geometry

10

200

Symmetric

4.6

1.0

30.0

Symmetric

10.0 15.0 48.7

2.3 3.3 9.6

30.0 30.0 200

Symmetric

52.5

14.0

200

Symmetric

48.7 22.9

9.6 7.3

200 –

Symmetric

43.4

12.7

200

Symmetric

– (a) Milli-Q water solution of NaCl (5.0 wt%) and glycerol with a weight 1.8 ratio of 1:3 (b) Decane solution of CR-310 (3.0 wt%) (a) Silicon oil (KF96-50) (b) Milli-Q water with 1.0 wt% SDS –

26.7

6.6

100

Symmetric

(a) Soybean oil (b) Milli-Q water with 1.0 wt% SDS

–

40.8

10.8

200

Symmetric

(a) Soybean oil (b) Demineralized water with 1.0 wt% SDS

–

10

30 Asymmetric (a) MCT (medium-chain fatty acid triglyceride) (microslot) (b) Milli-Q water with 2.0 wt% Tween20 70 (channel) 30 Asymmetric (a) Soybean oil (b) Milli-Q water with 0.5 wt% SDS (microslot) 70 (channel)

50

10 D = 10

c

–

40

D =10

D

(a) Refined Soybean oil (b) 25 mM NaCl solution containing 0.45% w/w BSA. (a) Refined soybean oil (b) 1 wt% SDS in Milli-Q

Type

d

a

Pressure Droplet (KPa) Diameter (lm)

Lsta Sstb Dstc (lm) (lm) (lm)

50

b

Dispersed phase (a) Continuous phase (b)

44.1

13.1 6.7 4.6 –

(a) Refined soybean oil (b) Demineralized water with 1.0 wt% SDS (a) Refined soybean oil (b) De-ionized water with 1.0 wt% Tween-20.

CV (%)

–

–

–

References

5.4 Saito et al. (2005)

4.4

5.5 Kobayashi et al. (2008b) 6.7 3.9 9.8 2.7 39.1 2.5 Kobayashi et al. (2003) 48.9 2.1 Kobayashi and Nakajima (2002a) 38.1 1.9 25.6 3.2 Kobayashi et al. (2008a) 49.8 1.7 Kobayashi et al. (2005b) 31–32 9–10 (Kobayashi et al. (2005c) 41.9 1.9 Kobayashi et al. (2004b) 27.1–27.6 – Vladisavljevic et al. (2010)

26.5

3–4

Vladisavljevic et al. (2008)

Longer length of straight-through channel (symmetric)/microslot (asymmetric). Shorter length of straight-through channel (symmetric)/microslot (asymmetric). Depth of straight-through channel (symmetric)/microslot (asymmetric). Diameter of cylindrical channel (asymmetric).

Table A3 Plateau dimensions and resulting droplet diameters under various experimental conditions for EDGE systems. Plateau dimensions

a b c d

Dispersed phase (a) Continuous phase (b)

Pressure (mbar) Droplet CV (%) References Diameter (lm)

Lpa (lm) Wpb (lm)

Hpc (lm) Npd ()

200 200 200 200 200

1.2 2.6 1.2 1.2 1.2

1 1 196 14 112

(a) Hexadecane (b) Milli-Q (1% SDS)

200

500 500 500 9500 200–1200 (triangular shaped) 500

1.2

196

200

500

1.2

–

200

500

1.2

–

(a) Sunflower oil (b) WPC solution (6% w/v) 310 or skim milk 400 (a) water-in-sunflower oil (10% w/w PGPR) (b) WPC solution (6% w/v) or skim milk (a) air (b) WPC solution (6% w/v) or skim milk 1000

Length of the plateau. Width of the plateau. Height of the plateau. Total number of plateaus.

(a) Hexadecane (b) Milli-Q (1% SDS)

175 80 180 160 160

7.1 14.6 7.0 7.0 7.0

5 5 4–5 5 5

van Dijke et al. (2010d)

7.2

11.8

van Dijke et al. (2010c)

7.4

6

30

van Dijke et al. (2009)

A.A. Maan et al. / Journal of Food Engineering 107 (2011) 334–346

Fig. B1. Schematic cross section of a microchannel showing length of neck (Ln) and length of channel (Lch) (van Dijke et al. (2008)).

and for this surface modification will be needed. Also, for industrial application, the material to be used for the construction of the devices needs to be (re-) considered, and it is expected that the current Si-based microchips will not be suitable for this. Most probably, polymer-based microchips, or even metal ones if they allow structure formation at the scale needed in microstructures, will be preferred. Which of these factors will be the determining one for industrial application is still not clear. But it is sure that there is still sufficient room for optimization and maturation of spontaneous emulsification technology to make it an attractive alternative for classic emulsification techniques. Appendix A See Tables A1–A3. Appendix B Flux through the entire system is calculated with the HagenPoisseuille equation, given as (Fig. B1)

Ut ¼

  dV R4 2r Papp  ¼P c dt 8gLch Rd;t

where, Ut is the total flux, Rc is the radius of channel, Lch is the length of channel (see Fig. 9) g is the fluid viscosity, Papp is the applied pressure, r is the interfacial tension and Rd is the droplet radius. Flux through the channel to the neck is calculated as

Uch;t ¼

  dV R4 r cos h ¼P c Papp  dt Rc 8gLch

and the flux through neck to the droplet is calculated as

Un;t ¼

dV R4 ¼P c dt 8gLn



r cos h Rc



2r Rt;d



The droplet breakup criterion is defined as

Un;t iUch;t References Abrahamse, A.J., van Lierop, R., van der Sman, R.G.M., van der Padt, A., Boom, R.M., 2002. Analysis of droplet formation and interactions during cross-flow membrane emulsification. Journal of Membrane Science 204, 125–137. Anna, S.L., Bontoux, N., Stone, H.A., 2003. Formation of dispersions using ‘‘flow focusing’’ in microchannels. Applied Physics Letters 82 (3), 364–366.

345

Arafat, A., Giesbers, M., Rosso, M., Sudholter, E.J.R., Schroën, K., White, R.G., Yang, L., Linford, M.R., Zuilhof, H., 2007. Covalent biofunctionalization of silicon nitride surfaces. Langmuir 23(11), 6233–6244. Arafat, A., Schroën, K., de Smet, L.C.P.M., Sudholter, E.J.R., Zuilhof, H., 2004. TailorMade functionalization of silicon nitride surfaces. Journal of the American Chemical Society 126 (28), 8600–8601. Arshady, R., 1993. Microcapsules for food. Journal of Microencapsulation 10 (4), 413–435. Bor, F.C., Calvin, G.J., 1997. Separation of submicron pharmaceutic emulsions with centrifugal split-flow thin (SPLITT) fractionation. Journal of Microcolumn Separations 9 (3), 205–211. Charcosset, C., 2009. Preparation of emulsions and particles by membrane emulsification for the food processing industry. Journal of Food Engineering 92 (3), 241–249. Charcosset, C., Limayem, I., Fessi, H., 2004. The membrane emulsification process-a review. Journal of Chemical Technology & Biotechnology 79 (3), 209–218. Forssell, P., Partanen, R., Poutanen, K., 2006. Microencapsulation – Better performance of food ingredients. Food Science and Technology 20 (3), 18–20. Friberg, S.E., Larsson, K., 1997. Food Emulsions. Marcel Dekker, Inc, USA. Fujiu, K.B., Kobayashi, I., Uemura, K., Nakajima, M., 2010. Temperature effect on microchannel oil-in-water emulsification Microfluidics and Nanofluidics 10.1007/s10404-010-0708-y. Huisman, I.H., Trägårdh, G., Trägårdh, C., Pihlajamäki, A., 1998. Determining the zeta-potential of ceramic microfiltration membranes using the electroviscous effect. Journal of Membrane Science 147 (2), 187–194. Iwamoto, S., Nakagawa, K., Sugiura, S., Nakajima, M., 2002. Preparation of gelatin microbeads with a narrow size distribution using microchannel emulsification. AAPS Pharmacology and Science Technology 3 (3). Joscelyne, S.M., Trägårdh, G., 2000. Membrane emulsification – a literature review. Journal of Membrane Science 169 (1), 107–117. Kawakatsu, T., Trägårdh, G., Kikuchi, Y., Nakajima, M., Komori, H., Yonemoto, T., 2000. Effect of microchannel structure on droplet size during crossflow microchannel emulsification. Journal of Surfactants and Detergents 3 (3), 295–302. Kawakatsu, T., Trägårdh, G., Trägårdh, C., 2001a. Production of W/O/W emulsions and S/O/W pectin microcapsules by microchannel emulsification. Colloids and Surfaces A: Physicochemical and Engineering Aspects 189 (1-3), 257–264. Kawakatsu, T., Trägårdh, G., Trägårdh, C., Nakajima, M., Oda, N., Yonemoto, T., 2001b. The effect of the hydrophobicity of microchannels and components in water and oil phases on droplet formation in microchannel water-in-oil emulsification. Colloids and Surfaces A: Physicochemical and Engineering Aspects 179 (1), 29–37. Kobayashi, I., Hirose, S., Katoh, T., Zhang, Y., Uemura, K., Nakajima, M., 2008a. Highaspect-ratio through-hole array microfabricated in a PMMA plate for monodisperse emulsion production. Microsystem Technologies 14 (9), 1349– 1357. Kobayashi, I., Lou, X., Mukataka, S., Nakajima, M., 2005a. Preparation of monodisperse water-in-oil-in-water emulsions using microfluidization and straight-through microchannel emulsification. Journal of the American Oil Chemists’ Society 82 (1), 65–71. Kobayashi, I., Mukataka, S., Nakajima, M., 2004a. CFD simulations and analysis of emulsion droplet formation from straight-through microchannels. Langmuir 20, 9868–9877. Kobayashi, I., Mukataka, S., Nakajima, M., 2004b. Effect of slot aspect ratio on droplet formation from silicon straight-through microchannels. Journal of Colloid and Interface Science 279 (1), 277–280. Kobayashi, I., Mukataka, S., Nakajima, M., 2005b. Effects of type and physical properties of oil phase on oil-in-water emulsion droplet formation in straightthrough microchannel emulsification, experimental and CFD studies. Langmuir 21(13), 5722–5730. Kobayashi, I., Mukataka, S., Nakajima, M., 2005c. Production of monodisperse oil-inwater emulsions using a large silicon straight-through microchannel plate. Industrial & Engineering Chemistry Research 44 (15), 5852–5856. Kobayashi, I., Nakajima, M., 2002a. Effect of emulsifiers on the preparation of foodgrade oil-in-water emulsions using a straight-through extrusion filter. European Journal of Lipid Science and Technology 104 (11), 720–727. Kobayashi, I., nakajima, M., Chun, K., kikuchi, Y., Fujita, H., 2002b. Silicon array of elongated through-holes for monodisperse emulsion droplets. AICh Journal 48 (8), 1639–1644. Kobayashi, I., Nakajima, M., Mukataka, S., 2003. Preparation characteristics of oil-inwater emulsions using differently charged surfactants in straight-through microchannel emulsification. Colloids and Surfaces A: Physicochemical and Engineering Aspects 229 (1-3), 33–41. Kobayashi, I., Nakajima, M., Nabetani, H., Kikuchi, Y., Shohno, A., Satoh, K., 2001. Preparation of micron-scale monodisperse oil-in-water microspheres by microchannel emulsification. Journal of the American Oil Chemists’ Society 78 (8), 797–802. Kobayashi, I., Takano, T., Maeda, R., Wada, Y., Uemura, K., Nakajima, M., 2008b. Straight-through microchannel devices for generating monodisperse emulsion droplets several microns in size. Microfluidics and Nanofluidics 4 (3), 167–177. Kobayashi, I., Uemura, K., Nakajima, M., 2008c. Generation characteristics of highly uniform nonspherical droplets of soybean oil using microchannel array devices. Food Biophysics 3 (2), 132–139. Kobayashi, I., Wada, Y., Uemura, K., Nakajima, M., 2010. Microchannel emulsification for mass production of uniform fine droplets: integration of microchannel arrays on a chip. Microfluidics and Nanofluidics 8, 255–262.

346

A.A. Maan et al. / Journal of Food Engineering 107 (2011) 334–346

Lambrich, I., Schubert, H., 2005. Emulsification using microporous systems. Journal of Membrane Science 257, 76–84. Liu, H., Nakajima, M., Kimura, T., 2004. Production of monodispersed water-in-oil emulsions using polymer microchannels. Journal of the American Oil Chemists’ Society 81 (7), 705–711. Liu, H., Nakajima, M., Nishi, T., Kimura, T., 2005a. Effect of channel structure on preparation of a water-in-oil emulsion by polymer microchannels. European Journal of Lipid Science and Technology 107 (7-8), 481–487. Liu, H., Nakajima, M., Nishi, T., Kimura, T., 2005b. Hydrophilic modification of polymer microchannel for preparation of oil-in-water emulsions. Nippon Shokuhin Kagaku Kogaku Kaishi 52 (12), 599–604. Nakagawa, K., Iwamoto, S., Nakajima, M., Shono, A., Satoh, K., 2004. Microchannel emulsification using gelatin and surfactant-free coacervate microencapsulation. Journal of Colloid and Interface Science 278 (1), 198–205. Nakashima, T., Shimizu, M., Kukizaki, M., 1991. Membrane emulsification by microporous glass. Key Engineering Materials (61/62), 513. Neves, M., Ribeiro, H., Kobayashi, I., Nakajima, M., 2008. Encapsulation of lipophilic bioactive molecules by microchannel emulsification. Food Biophysics 3 (2), 126–131. Romoscanu, A.I., Fenollosa, A., Acquistapace, S., Gunes, D., Deuchande, T., Clausen, P., Mezzenga, R., Nyden, M., Zick, K., Hughes, E., 2010. Structure, diffusion and permeability of protein-stabilized monodispersed oil-in-water emulsions and their gels: A self diffusion NMR study. Langmuir 26(9), 6184–6192. Rosso, M., Giesbers, M., Arafat, A., Schroën, K., Zuilhof, H., 2009. Covalently attached organic monolayers on SiC and SixN 4 surfaces: Formation using UV light at room temperature. Langmuir 25(4), 2172–2180. Sagis, L.M.C., de Ruiter, R., Miranda, F.J.R., de Ruiter, J., Schroën, K., van Aelst, A.C., Kieft, H., Boom, R., van der Linden, E., 2008. Polymer microcapsules with a fiberreinforced nanocomposite shell. Langmuir 24, 1608–1612. Saito, M., Yin, L.J., Kobayashi, I., Nakajima, M., 2005. Preparation characteristics of monodispersed oil-in-water emulsions with large particles stabilized by proteins in straight-through microchannel emulsification. Food Hydrocolloids 19 (4), 745–751. Saito, M., Yin, L.J., Kobayashi, I., Nakajima, M., 2006. Comparison of stability of bovine serum albumin-stabilized emulsions prepared by microchannel emulsification and homogenization. Food Hydrocolloids 20 (7), 1020–1028. Steegmans, M.L.J., 2009. emulsification in microfluidic Y- and T-junctions. Ph.D thesis, Wageningen Uivresity (WUR). Sugiura, S., Nakajima, M., Itou, H., Seki, M., 2001a. Synthesis of polymeric microspheres with narrow size distributions employing microchannel emulsification. Macromolecular Rapid Communications 22 (10), 773–778. Sugiura, S., Nakajima, M., Iwamoto, S., Seki, M., 2001b. Interfacial tension driven monodispersed droplet formation from microfabricated channel array. Langmuir 17(18), 5562–5566. Sugiura, S., Nakajima, M., Kumazawa, N., Iwamoto, S., Seki, M., 2002a. Characterization of spontaneous transformation-based droplet formation during microchannel emulsification. Journal of Physical Chemistry B 106 (36), 9405–9409. Sugiura, S., Nakajima, M., Oda, T., Satake, M., Seki, M., 2004a. Effect of interfacial tension on the dynamic behavior of droplet formation during microchannel emulsification. Journal of Colloid and Interface Science 269 (1), 178–185. Sugiura, S., Nakajima, M., Seki, M., 2002b. Effect of Channel Structure on Microchannel Emulsification. Langmuir 18(15), 5708–5712. Sugiura, S., Nakajima, M., Seki, M., 2002c. Prediction of droplet diameter for microchannel emulsification. Langmuir 18(10), 3854–3859. Sugiura, S., Nakajima, M., Seki, M., 2002d. Preparation of monodispersed emulsion with large droplets using microchannel emulsification. Journal of the American Oil Chemists’ Society 79 (5), 515–519.

Sugiura, S., Nakajima, M., Seki, M., 2002e. Preparation of monodispersed polymeric microspheres over 50 lm employing microchannel emulsification. Industrial & Engineering Chemistry Research 41 (16), 4043–4047. Sugiura, S., Nakajima, M., Tong, J., Nabetani, H., Seki, M., 2000. Preparation of monodispersed solid lipid microspheres using a microchannel emulsification technique. Journal of Colloid and Interface Science 227 (1), 95–103. Sugiura, S., Nakajima, M., Yamamoto, K., Iwamoto, S., Oda, T., Satake, M., Seki, M., 2004b. Preparation characteristics of water-in-oil-in-water multiple emulsions using microchannel emulsification. Journal of Colloid and Interface Science 270 (1), 221–228. Tan, Y.C., Cristini, V., Lee, A.P., 2006. Monodispersed microfluidic droplet generation by shear focusing microfluidic device. Sensors and Actuators B: Chemical 114 (1), 350–356. Tan, Y.C., Fisher, J.S., Lee, A.I., Cristini, V., Lee, A.P., 2004. Design of microfluidic channel geometries for the control of droplet volume, chemical concentration, and sorting. Lab on a Chip 4 (4), 292–298. Tong, J., Nakajima, M., Nabetani, H., 2002. Preparation of phospholipid oil-in-water microspheres by microchannel emulsification technique. European Journal of Lipid Science and Technology 104 (4), 216–221. Tong, J., Nakajima, M., Nabetani, H., Kikuchi, Y., 2000. Surfactant effect on production of monodispersed microspheres by microchannel emulsification method. Journal of Surfactants and Detergents 3 (3), 285–293. Umbanhowar, P.B., Prasad, V., Weitz, D.A., 2000. Monodisperse Emulsion Generation via Drop Break Off in a Coflowing Stream. Langmuir 16(2), 347–351. Utada, A.S., Chu, L.Y., Fernandez-Nieves, A., Link, D.R., Weitz, D.A., 2007. Dripping, jetting, drops, and wetting: the magic of microfluidics. MRS Bulletin 32, 702– 708. van der Graaf, S., Schroën, C.G.P.H., Boom, R.M., 2005. Preparation of double emulsions by membrane emulsification–A review. Journal of Membrane Science 251 (1-2), 7–15. van Dijke, K.C., de Ruiter, R., Schroën, K., Boom, R.M., 2010a. The mechanism of droplet formation in microfluidic EDGE systems. Soft Matter 6 (2), 321–330. van Dijke, K.C., Kobayashi, I., Schroën, K., Uemura, K., Nakajima, M., Boom, R.M., 2010b. Effect of viscosities of dispersed and continuous phases in microchannel oil-in-water emulsification. Microfluidics and Nanofluidics 9, 77–85. van Dijke, K.C., Schroën, K., van der Padt, A., Boom, R.M., 2010c. EDGE emulsification for food-grade dispersions. Journal of Food Engineering 97 (3), 348–354. van Dijke, K.C., Schroën, K.C.G.P.H., Boom, R.M., 2008. Microchannel Emulsification: From Computational Fluid Dynamics to Predictive Analytical Model. Langmuir 24(18), 10107–10115. van Dijke, K.C., Veldhuis, G., Schroen, K., Boom, R.M., 2009. Parallelized edge-based droplet generation (EDGE) devices. Lab on a Chip 9 (19), 2824–2830. van Dijke, K.C., Veldhuis, G., Schroën, K., Boom, R.M., 2010d. Simultaneous formation of many droplets in a single microfluidic droplet formation unit. AIChE Journal 56 (3), 833–836. Vilstrup, P., 2001. Microencapsulation of Food Ingredients. Leatherhead publishing. Vladisavljevic, G.T., Kobayashi, I., Nakajima, M., 2008. Generation of highly uniform droplets using asymmetric microchannels fabricated on a single crystal silicon plate: Effect of emulsifier and oil types. Powder Technology 183 (1), 37–45. Vladisavljevic´, G.T., Kobayashi, I., Nakajima, M., 2010. Effect of dispersed phase viscosity on maximum droplet generation frequency in microchannel emulsification using asymmetric straight-through channels. Microfluidics and Nanofluidics 10.1007/s10404-010-0750-9. Yasuno, M., Sugiura, S., Iwamoto, S., Nakajima, M., Shono, A., Satoh, K., 2004. Monodispersed microbubble formation using microchannel technique. AIChE Journal 50 (12), 3227–3233.