Synthesis of polymer particles and capsules employing microfluidic techniques

1 Introduction

Apart from the significant achievements that microreactor technology has made in chemical synthesis and chemical engineering, specially developed microfluidic structures have also gained great importance in generating fluid segments in the form of droplets and bubbles. The size of the fluid segments formed and the frequency of segment formation can be controlled very precisely through the selected flow conditions, channel geometries, and other process parameters. On the basis of these droplets, microscale polymer particles can be synthesized by using appropriate monomers or prepolymers as the dispersed phase. The subsequent polymerization is conducted under precisely controlled process conditions. To initiate the polymerization reaction, either heat or UV light can be accurately introduced into the microfluidic reactor, and heat arising from exothermic polymerization reactions can be removed instantaneously due to the huge internal surface area of microchannels.

In comparison to conventional mechanically agitated batch processes, polymer particles of well-defined shape and size can be produced continuously, providing very narrow particle size distributions.

In this mini review, we provide a short summary of microfluidic techniques that have been employed for droplet formation and the subsequent synthesis of polymer particles including the more sophisticated synthesis of polymer capsules. Finally, we consider major challenges, which have to be overcome in order to establish these new microfluidic techniques as a competitive alternative to conventional production processes.

2 Droplet formation

For microfluidic droplet formation, two different types of devices are generally used – capillary-based devices and microstructured chip devices. The former are mostly self-constructed arrangements of coaxially nested capillaries of different sizes, which usually have circular cross-sections. The capillaries are mainly made of glass due to the possibility of precisely adjusting the diameter and, particularly, the wall thickness of their tips. In order to handle the capillaries in a rugged setup and allow the attachment of connecting tubes, the microfluidic arrangements are often cast in different acrylic or silicone resins. Different geometries of capillary-based devices for droplet generation have been achieved (Figure 1, left). For example, in the coflow arrangement, one fluid flows through the inner capillary, while a second fluid is passed through the outer one flowing in the same direction. The droplets are formed at the tip of the inner capillary due to shear forces and have diameters that are usually larger than the orifice of the tip. Where a countercurrent flow is realized forcing the two fluids through a tapered end of the inner capillary, an additional advantageous effect, usually referred to as “flow focusing,” can be exploited. In this case, by applying a small orifice, smaller droplets can be formed, and the droplet size can be adjusted in a wider flow rate range.

Figure 1

Schematic drawings of different microdevices for droplet formation; (left) capillary based droplet devices, (right) planar microstructured droplet devices.

In planar microstructured devices, which typically have noncircular cross-sections, three basically different geometries for droplet formation with different variations have been realized (Figure 1, right). The simplest one is a T-junction where the dispersed phase is dosed perpendicularly to the main flow. At low flow rates, an emerging droplet grows until it occupies nearly the whole cross-section of the channel and then breaks up due to squeezing by the continuous phase and flows along the main channel. This is usually referred to as a squeezing [1] or slugging [2] regime (Figure 2). In the dripping regime, at higher flow rates, the droplet is pinched off earlier due to higher shear forces. In this way, droplets can be generated, which are significantly smaller than the channel diameter. This regime is, therefore, more important for polymer particle synthesis, as small droplets are usually desired, and further processing is easier if droplets have no contact to the channel wall. Apart from squeezing and dripping, there are other flow regimes like the jetting regime where elongated threads of the dispersed phase are formed, and the droplets break up somewhere downstream. The jetting regime is a common side effect of droplet formation processes and can, thus, often not be avoided to a certain extent. This results in a broader droplet size distribution due to less regular droplet break up and the formation of small secondary “satellite” droplets.

Figure 2

Different modes of droplet formation in a T-junction, (A) squeezing regime, (B) dripping regime, (C) jetting regime.

A basically different geometry is the coflow arrangement in two intersected microchannels (Figure 1, right). Droplet formation occurs by the confinement and shearing of the dispersed phase caused by the coflowing continuous streams in the perpendicular outer channels. In general, the modes of droplet formation are the same as in T-junctions, whereas the mostly desired dripping regime is obtained in a wider flow rate range. This can be further improved by additionally focusing the flow by using a narrow orifice in the outlet channel. Consequently, such devices are called flow-focusing devices, and they are most widely used as they provide the best droplet formation characteristics.

The design of microfluidic droplet generators usually aims at the deliberate exploitation of physical forces influencing droplet formation, namely, viscous shear and interfacial forces. Their interaction can be expressed by the capillary number Ca=μU/γ, where μ is the viscosity, U is the flow velocity, and γ is the interfacial tension. Accordingly, droplet formation and the size of the droplets formed are greatly influenced by both material properties and flow conditions such as total flow rate and the flow rate ratio of the fluids involved. The impact of these properties and parameters has been experimentally analyzed in numerous studies [1–8]. An extensive summary of all findings is not possible here, but one can conclude from these studies that droplet size can best be controlled and manipulated by the flow rate and the flow rate ratio of the two liquid phases. For example, in the squeezing, dripping, and – to a certain extent – also in the jetting regime, smaller droplets are usually obtained at a higher flow rate. However, in flow-focusing devices, the size of the droplets is also strongly influenced by the diameter of the orifice.

3 Polymer particle synthesis

Syntheses of polymer particles in microfluidic devices are usually carried out in a single- or two-step process. For example, the one-step synthesis of acrylate particles was realized by a projection photolithography technique where a monomer solution of multifunctional acrylates, e.g., poly(ethylene glycol) diacrylate, trimethylpropane triacrylate, or 1,6-hexanediol diacrylate, containing a photoinitiator, was irradiated in situ with UV light through a patterned mask [9]. Only the mask-defined areas are crosslinked by exposure to pulsed UV light and due to rapid polymerization kinetics solid structures are formed quickly. Various shapes, e.g., polygonal, round, and curved, of particles with characteristic diameters in the range of 10–50 μm were obtained by the use of differently patterned masks. Also Janus (two-faced) type particles with asymmetric functionalities, e.g., hydrophilic and lipophilic properties, were synthesized by realizing a laminar coflow and subsequent UV irradiation in a microchannel [10].

The two-step syntheses of polymer particles (Figure 3) are multiphase processes based on emulsification, i.e., droplet formation by the previously described microfluidic techniques and subsequent solidification. In the first applications, the solidification of monomer droplets (e.g., divinylbenzene or 1,6-hexanediol diacrylate) was realized outside of the microfluidic device by collecting the particles in a beaker followed by immediate curing (started either by thermal or photo initiation) (Figure 3, middle) [11, 12]. Strictly speaking, these are not continuous microfluidic processes in the true sense. Although droplet formation or emulsification was performed continuously, the polymerization step more closely resembles a batch process. However, it could be turned to continuous operation by applying adequately sized continuous-stirred tank reactors, which can also be capable for photochemical reactions. There are several advantages associated with this off-chip curing methodology, e.g., the curing time can be easily adjusted and adapted to the necessary reaction time and also the clogging of microchannels can easily be avoided. On the other hand, the main disadvantage of this methodology is the coalescence of emulsion droplets, which occurs during their transfer from the microfluidic device to the batch collector and during the batch polymerization itself. This usually results in a significant broadening of the size distribution of polymer particles compared to the previously formed emulsion droplets.

Figure 3

Different processes for polymer particle polymerization; (top) continuous on-chip curing by UV initiation, (middle) curing outside of the microfluidic device by thermal or UV initiation, (bottom) continuous on-chip precuring by UV initiation and off-chip postpolymerization.

Continuous polymerization was also successfully realized by droplet formation and the direct on-chip UV curing of the precursor droplets in an extended microchannel (Figure 3, top) [13–16]. The coalescence of precursor droplets can be counteracted as a certain distance between adjacent droplets can be adjusted and maintained in the microchannel by choosing an appropriate flow rate of the continuous phase. Moreover, the narrow size distribution of the precursor droplets is usually maintained during the polymerization. Nevertheless, there are also certain limitations for this methodology. As the entire polymerization is conducted on-chip, the reaction time directly determines the required residence time in the microfluidic device and, thus, the inner volume of the chip as well as the overall flow rate. As described above, the flow rate cannot be reduced indefinitely to provide longer residence times as the droplet formation in microchannels is also highly dependent on the flow rate. Consequently, for realizing a certain droplet size, the only possibility for adjusting the reaction time is by implementing the right length, i.e., volume of the microchannel, or conduct the reaction in stopped flow. In stopped flow, even the polymerization of tube-confined monomer droplets leading to rather big crosslinked polystyrene particles by thermal curing for about 12 h was performed [17]. Anyway, for continuous processing, the microfluidic devices have to be customized for a certain polymerization reaction, and the maximum reaction time is very limited. Free radical polymerizations, e.g., by the application of multifunctional acrylates as monomers, are, therefore, promising and already applied candidates for on-chip particle formations due to their high polymerization rate.

Nevertheless, the on-chip conversion is often not high enough, which makes external batch-wise postpolymerization of partly on-chip polymerized particles necessary (Figure 3, bottom). Another limitation associated with on-chip polymerization is the risk of clogging of the microchannels. This can simply be due to the aggregation of particles, but can also result from uncontrolled polymerization of the monomer solution. In order to achieve high polymerization rates on-chip, fairly reactive monomer solutions like multifunctional acrylates with a high content of initiators (in the range of 3.5–4.5 wt%) are usually used [18]. Minor fluctuations of flow, disruption of droplet formation, and also the unexpected irradiation of feeding channels can quickly result in blocked channels, which makes this methodology less robust than off-chip polymerization.

By using the above-described microfluidic techniques, numerous types of polymer particles have been produced by different research groups. Among the polymer systems investigated are divinylbenzene [7, 11], acrylates (e.g., hexanediol diacrylate [12], ethylene glycol dimethacrylate [15, 19], tripropylene glycol diacrylate [20], pentaerythritol triacrylate [20]), and different microgels [21–23]. By applying a coflow of two monomer solutions, Janus-type particles were also synthesized (Figure 4) [12, 14, 24]. Composite particles can also be obtained by mixing an additive with the monomer solution prior to emulsification. Particles loaded with different semiconductor or magnetic nanoparticles were produced in this way [21, 25, 26]. In a recent work [27], polyacrylamide particles incorporating silver nanoparticles were produced as on-chip probes for microfluidic surface-enhanced Raman spectroscopy (SERS). Finally, the functionalization of polymer particles can also be realized in microfluidic processes, for example, by synthesizing porous particles like molecularly imprinted polymer (MIP) beads, which are used for the trace detection of analytes and for chromatographic purposes [19, 28, 29].

Figure 4

Microscope images of Janus particles (A–C) and particles with ternary structures (D). Bright phase is methacryloxypropyl dimethylsiloxane and 4±1 wt% of 1-hydroxycyclohexyl phenyl ketone, dark phase is a mixture of pentaerythritol triacrylate (45.0 wt%), poly(ethylene glycol) diacrylate (45.0 wt%), and acrylic acid (5.0 wt%) plus 4±1 wt% of 1-hydroxycyclohexyl phenyl ketone. Small images in (A–D) are fluorescence images of the corresponding particles. (E) Size distribution of particles shown in (A). CV=1.8%. Reprinted with permission from [14]. Copyright (2006) American Chemical Society.

4 Microcapsule synthesis

Microcapsules are polymer or gel microparticles containing a well-defined core material, which can be gaseous, liquid, or solid. At present, several conventional encapsulation techniques are available. Their main limitation is the size distributions that can be achieved, concerning both absolute capsule size and capsule shell thickness, which both should preferably be narrower for many applications, e.g., advanced drug delivery (as drug release is proportional to microparticle size) [30].

Microfluidic polymer capsule syntheses have been realized in analogy to microfluidic polymer particle syntheses. They are also based on an initial continuous droplet, i.e., emulsion, formation followed by different encapsulation methods on the basis of interfacial polymerization, solvent removal, polymerization of shells, or gelation.

Based on simple oil-in-water (O/W) or water-in-oil (W/O) emulsions, microencapsulation can be achieved by interfacial polymerization. In this case, each phase contains at least one of the necessary polymerization reaction components. The reaction takes place after droplet formation at the newly established interface between the two phases (Figure 5A). An example for this method is the formation of nylon-6,6-coated aqueous droplets [31]. In this process, aqueous droplets containing 1,6-diaminohexane were formed within a continuous hexadecane phase by the use of a microfluidic flow-focusing device. Through an additional feed, a solution of adipoyl chloride in dichloroethane and hexadecane was introduced into the outlet stream. As soon as the adipoyl chloride struck the aqueous droplets, interfacial polymerization occurred rapidly by polycondensation [31]. In a similar manner, polyamide capsules were also formed by using a simple T-junction arrangement for droplet formation and subsequent contacting of polyethylene diamine with sebacoyl chloride and benzene tricarboxylic acid [32]. As an alternative to a contact reaction between two reagents, Choi et al. [33] prepared monodisperse poly(N-isopropylacrylamide) (PNIPAM) microcapsules in a microfluidic system by forming droplets of an aqueous monomer solution (NIPAM) in an immiscible continuous phase. The continuous phase contained a photoinitiator, which initiated polymerization at the interface upon irradiation, thus, forming hollow microcapsules. In each case, it remains difficult to control the shell thickness of capsules in an appropriate manner by applying interfacial polymerization. Own studies on the synthesis of polyamide capsules have shown that the mechanical and long-term stability of the formed capsules are often insufficient for technical applications.

Figure 5

Schemes depicting microfluidic capsules syntheses by (A) interfacial polymerization, (B) initial formation of multiple emulsions with a liquid core-shell architecture, and subsequent polymerization or gelation of the shell.

A further method for the formation of polymer capsules in microfluidic devices is based on the initial formation of multiple emulsions with a liquid core-shell architecture and subsequent polymerization or gelation of these shells (Figure 5B). Multiple emulsions can be formed either in capillary or planar microfluidic devices. Compared to the formation of single emulsions, the fluidic design of these devices is more complex as more feeding channels are needed. As an example, Figure 6 shows the formation of different multiple emulsions (W/O/W) in the same microfluidic glass device with two intersections. The inner liquid phase was water; toluene with 1% of Span 20 was used as the shell phase, whereas water with 4% polyvinyl alcohol (PVA) was used as the continuous phase. Where a monomer solution like a multifunctional acrylate was used as shell material, microcapsules could be formed by polymerization similarly to the previously described polymer particle syntheses. The outer droplet (and future capsule) size, the size of the inner droplets, the number of encapsulated inner droplets, as well as the shell thickness of the capsules can be controlled accurately by the overall flow rate and the flow rate ratios of the phases involved. In order to allow the formation of sufficiently stable multiple emulsions, it is essential to choose suitable liquid phase materials and to tune the surface energies by the application of surfactants. The stability of multiple emulsions is determined by multiple factors, e.g., the balance of Laplace and osmotic pressures between the distinct droplets, viscosity ratio of the involved phases, and interactions of the surfactants at the different interfaces [34, 35]. These different factors can be taken in account and manipulated in a model system like the one shown in Figure 6, but as soon as the liquid phases are not freely selectable, due to the desired capsule properties and applications, it can become much more challenging or even impossible to establish stable multiple emulsion as the basis for subsequent microcapsule formation processes. Moreover, the material properties and, in particular, the wetting behavior of microchannels and capillaries may also have a significant influence on a successful emulsion formation. For example, O/W emulsions are best formed in channels with hydrophilic surface properties, and hydrophobic channels are preferably used for W/O emulsions. As a result, it is difficult to form O/W/O or W/O/W emulsions in a microfluidic device exhibiting the same wetting behavior in all channels. Further aspects of different wetting behavior in microfluidic devices will be discussed in the following chapter.

Figure 6

Microscope images of exemplary multiple W/O/W emulsions with a different number of core droplets formed by variation of flow rate ratios of continuous and dispersed phase in a microfluidic glass device with a structure similar to Figure 5B (channel diameter of outlet channel: 500 μm).

Despite all the technical challenges described, the successful synthesis of microcapsules based on the shell solidification method has already been reported by different research groups [36–41]. Among the first were Utada et al., who synthesized microcapsules in a microfluidic capillary device by forming a double emulsion. Shell polymerization was realized either by incorporating a UV-curable commercial monomer mixture (Norland Optical Adhesive, NOA) or by applying a solvent evaporation technique to a diblock copolymer [poly(butyl acrylate)-b-poly(acrylic acid) (PBA-PAA)] dissolved in the liquid shell (Figure 7) [41]. The formation of capsules was proven by phase-contrast images and by mechanically crushing the capsules and, thus, revealing the capsular architecture. In other published works, microcapsule synthesis based on double emulsions was realized by forming shells incorporating tripropylene glycol diacrylate (TPGDA) and other acrylates [36, 40], poly(N-isopropylacrylamide) (PNIPAM) [37], polysolfone [42], alginate [38], and biodegradable polymers [39].

Figure 7

Core-shell structures fabricated from double emulsions generated in a capillary microfluidic device. (A) Optical photomicrograph of the water-in-oil-in-water double emulsion precursor to solid spheres. The oil consists of 70% NOA and 30% acetone. (B) Optical photomicrograph of rigid shells made by crosslinking the NOA by exposure to UV light. (C) SEM of the shells shown in (B) after they have been mechanically crushed. The scale bar in (C) also applies to (A) and (B). (D) Brightfield photomicrograph of the water-in-oil-in-water double emulsion precursor to a polymer vesicle. The oil phase consists of a mixture of toluene and tetrahydrofuran at 70/30 v/v with dissolved diblock copolymer (PBA-PAA) at 2% w/v. (E) Phase-contrast image of the diblock copolymer vesicle after evaporation of the organic solvents. (F) Phase-contrast image of the deflated vesicle after osmotic stress was applied through the addition of 0.1 m sucrose to the outer fluid. The scale bar in (F) also applies to (D) and (E). Reprinted with permission from [41]. Copyright (2005) American Association for the Advancement of Science.

5 Wetting behavior of microchannels

It was mentioned previously that the wettability of microfluidic channels is crucial for droplet formation, i.e., emulsification, and especially multiple emulsion formation. In this regard, glass capillary devices offer certain benefits. Owing to their circular shape, they provide three-dimensional fluidic structures that prevent unwanted wall contacts of fluid segments. For example, in capillary flow-focusing devices, droplets are released from a smaller capillary coaxially into a larger capillary without being target to the capillary wall. In contrast, it is more difficult to generate similar coaxial three-dimensional fluidic structures in planar microfluidic devices as this requires different channel depths. The microfabrication of such devices by commonly applied lithography, etching, or abrasive techniques becomes more complex. Consequently, in many planar microfluidic devices, the dispersed phase cannot be completely surrounded by the continuous phase during the formation of an emerging droplet and can, therefore, more easily stick to the channel wall instead of forming isolated droplets. It is, therefore, essential that the surface of the fluidic channel does not attract the formed droplet by exhibiting appropriate hydrophilic or hydrophobic surface properties according to the material properties of the droplet phase. For example, the surface properties of glass can be easily modified by chemical surface treatment, e.g., silanization reactions or sol-gel coatings. As microfluidic capillary devices usually consist of several (at least two) capillaries, each capillary surface can be treated separately in order to achieve the required wettability; the different capillaries are then assembled to the corresponding microfluidic device. In general, a similar approach can also be realized in planar microfluidic channel structures. Abate et al. [43] proposed a technique to pattern wettability using surface coating and flow confinement and W/O/W and O/W/O double emulsions formed in such tailored devices. Abate et al. used a sol-gel coating approach to polydimethylsiloxane (PDMS) devices, but other surface treatment methods can be applied in a similar way. The generalized principle of this local surface modification method is depicted in Figure 8. When applied to a simple channel crossing geometry, the chemically active coating fluid can be pumped into one inlet channel (left channel in Figure 8), whereas an inert fluid is introduced into the opposite inlet channel. The two fluids meet at the crossing, and a stable interface is formed. The upper and lower channels are open and act as outlets for both fluids. The flow rates need to be high enough to restrict diffusion across the interface and outlet channels.

Figure 8

Scheme for patterned surface treatment of glass microchannels using flow confinement. Liquids are pumped from both sides and leave the device through the upper and lower channel. By applying appropriate coating reagents, the left channel can be converted to hydrophobic, whereas the right channel remains hydrophilic.

We applied this technique in a slightly modified manner for patterning the wettability of a microstructured glass device consisting of two junctions (Figure 9) in order to form W/O/W double emulsions. As glass is naturally hydrophilic, the channel between the first and the second junction, where the W/O emulsion is formed and transported, had to be converted to hydrophobic. All microchannels were, therefore, initially converted to hydrophobic by conducting a surface silanization reaction with n-octadecyltrichlorosilane (OTS). By applying the described flow confinement technique, the hydrophobic coating was then removed locally in a second process step using a 10% sodium hydroxide solution in order to restore the hydrophilic glass surface at the water inlet and the double emulsion outlet channels. Figure 9 shows the W/O emulsion formation at the first junction before and after the local surface treatment (hydrophobization). Consequently, the formation of multiple emulsions in glass devices can be realized with a significantly higher flexibility concerning the application of different liquids and phases.

Figure 9

Scheme for multiple emulsion formation in a microfluidic glass device with patterned wettability. As a result of the hydrophobic coating in the middle section, stable W/O emulsions are formed. Owing to the hydrophilic glass surface in the outlet channel stable, W/O/W emulsions are generated in a subsequent step.

In the meantime, further techniques for achieving patterned surface treatment of microfluidic chip devices have been published, for example, based on microplasma treatment [44] and UV mask patterning [45]. All the techniques mentioned are very well suited for lab applications, but more difficulties are expected in the development of microfluidic bulk production processes for polymer particles and capsules, which will be discussed in the following section.

6 High throughput synthesis of particles and microcapsules

In the previous sections, it was shown that polymer microparticles and microcapsules can now be synthesized successfully by employing various microfluidic techniques. However, a common feature of all the examples mentioned is that they were realized in laboratory setups often consisting of tiny single-channel devices operated at very low throughputs, mostly in the μl/min range. Consequently, the productivity is very low, rarely exceeding the synthesis of a few hundred milligrams of product, which is needed for product characterization. In order to consider microfluidic techniques as an alternative production technology, their productivity must be significantly enhanced. The usual approach in microreaction technology for increasing throughput is the so-called numbering-up or equaling-up approach. Classical scale-up is not applicable here as the use of microfluidic structures is compulsory for the synthesis of polymer microparticles and microcapsules.

The numbering-up approach is based on the parallel repetition of microfluidic units, i.e., the parallelization of droplet formation units for the present applications [46]. An essential characteristic of this concept is that the microfluidic flow conditions known from a single unit are maintained despite the overall increase in throughput. Hence, it is crucial that the bulk feed stream is split uniformly into the multiple parallel units. However, slight flow variations between parallel microchannels or droplet-generating units of only a few μl/min may already lead to malfunction of the entire droplet formation process, which makes numbering-up an extremely difficult task. Nevertheless, in recent years, several studies on the development of parallelized microfluidic droplet formation devices and their application for polymer particle syntheses have been published. The most straightforward method is the combination of multiple droplet generators and required reaction channels within a single device or substrate [47, 48]. Special attention was paid to the design and geometry of the feeding and distribution channels in order to generate equal fluid resistance, i.e., pressure drop, in each channel and, thus, ensure equal flow rates. With such a parallelized system consisting of 15 droplet-generating units, Romanowsky et al. [47] are considering the production of 1.5 kg of double emulsion droplets per day. Nisisako et al. [49] realized a microfluidic platform with a circular arrangement of identical droplet generators (Figure 10). The evaluation of the flow velocity distribution in the multiple channels was carried out experimentally and by numerical simulation. In the microfluidic platform shown in Figure 10, they were able to combine 144 crossflow droplet generators or 72 droplet generators for Janus droplets or 32 triple emulsion generators. In the case of simple O/W emulsions, Nisisako et al. achieved an overall throughput of 180 ml/h, generating droplets with a mean diameter of 90.7 μm while maintaining a narrow droplet size distribution with a coefficient of variation (CV) of 2.2%. They also showed the successful formation of Janus droplets as well as double and triple emulsions, even though the throughput was lower for the more sophisticated droplet compositions. Finally, Kobayashi et al. [50, 51] chose a somewhat different approach by constructing membrane-like devices, which are composed of arrays of microfabricated holes for emulsification (Figure 11). In this way, they were able to realize up to 11,900 through-holes for droplet formation. The maximum flow rate of the dispersed phase was 1.5 ml/min with a very good droplet size uniformity. This value is significantly lower than the previously mentioned one, but it should be noted that Kobayashi et al. focused on the generation of very small droplets with diameters in the range of 10 μm.

Figure 10

Microfluidic platform for the generation of monodisperse O/W emulsions consisting of 144 droplet generators. (A, B) Optical microscope images showing arrayed O/W droplets formed alternately at parallelized double T-junctions. The disperse phase flow rate Qd was 180.0 ml/h, and the continuous phase flow rate Qc was 270.0 ml/h. (C, D) Optical microscope image and size distribution of the resulting droplets. The mean diameter Davg is 90.7 μm, with a CV of 2.2. (E) Evaluation of the disperse phase flow rate per channel (Qd,1) for different droplet generators. The horizontal axis shows the location of droplet generators in (A) (#0 to #22 from left to right). Open squares show the measured breakup frequency Fb,1 (average=856.3 1/s, CV=0.6%). Solid bars are Qd,1, calculated from the measured Fb,1 and the Davg (average=1.20 ml/h, CV=0.6%). The dashed line and dotted lines are estimates of Fb,1 (=888.8 1/s) and Qd,1 (=1.25 ml/h), calculated from the setting of the flow rate at the syringe pump. Reprinted with permission from [49]. Copyright (2012) The Royal Society of Chemistry.

Figure 11

Chip geometries for mass production of uniform fine droplets. (A) Schematic top view of a chip consisting of 14 arrays with microfabricated holes for emulsification. Solid circles denote the inlet through-holes for the continuous liquid phase and the outlet through-holes for the emulsion product. (B) Schematic top view of droplet generation via emulsification arrays on the chip. (C) Optical micrograph of generating fine oil droplets at a flow rate of the dispersed phase (Qd) of 0.5 ml/h. Reprinted with permission from [50]. Copyright (2009) Springer-Verlag.

7 Conclusion

Microfluidic technologies allow the continuous generation of a wide range of polymer microparticles and microcapsules by providing a precise control over process conditions and product properties such as particle size and size distribution. The main techniques based on droplet and emulsion formation as the initial process step were reviewed in this paper. Accordingly, an impressive variety of different particulate products has been synthesized in the past decade aiming at various potential applications, ranging from chemical and biological research, medical diagnostics and drug delivery to consumer products in food, cosmetic, and even textile industries. In order to allow widespread use and establish microfluidic processes in the production of microcapsules and microparticles, it will be essential to focus future work on the development of high-throughput processes. Some promising approaches have already been discussed in this minireview, but there are still much more innovative technical concepts to come.