Fig. 10.1
Illustration of microencapsulation by interfacial polymerization in O/W emulsion
Several methods are known to substantially produce monodisperse emulsions and therefore monodisperse microcapsules. Emulsification using Shirasu porous glass (SPG) is widely used for preparing microcapsules [17]; its use in bioscience and technology will be described in the next chapter. Monodisperse liquid droplets are also prepared by controlling the vibrational mode of the liquid column mechanically or electrically. By applying an alternating electric field on liquid flowing in air, the size of the resultant droplets is determined by the frequency of the electric field and surface tension. Monodisperse microcapsules with a standard deviation as low as 1.5 % were obtained by this method [18]. The coacervation method in microencapsulation utilizes the phase separation of polymer solutions, especially polyelectrolyte solutions. Droplets of the concentrated phase containing functional materials are fixed by different materials such as talc [Mg3Si4O10(OH)2] and extracted as a collection of microcapsules. The growth of liquid droplets in the process is induced by the chemical potential difference before and after the phase separation, and the growth rate depends on the diffusion constant of polymer molecules. Phase separation is conventionally classified into a nucleation and growth process and a spinodal decomposition process. The droplets produced by the former process have a large distribution because of the difference in the time of appearance of each nucleus; however, homogeneous and coherent droplet growth is realized in the latter process. Therefore, microencapsulation by means of spinodal decomposition can be used to design a preparation procedure for monodisperse microcapsules, and its optimum conditions are predicted by the free energy of the system [19].
10.2.2 Stability of Microcapsule Suspension
The stability of a suspension of microcapsules is mainly governed by the characteristics of their wall membrane. When the wall membrane has a net charge, the suspension is stable due to repulsive coulombic force. When it has no net charge, stability usually needs to be provided by a protective colloid due to the excluded volume effect. In the latter case, the suspension is quasi-stable, i.e., microcapsules finally aggregate with time, and the aggregates obtained can be redispersed by perturbations such as ultrasonic irradiation.
Microcapsules are also classified into two kinds depending on the cross-linking structure of their wall membrane: those with covalent cross-linkage and those with weakly interacting cross-linkages such as hydrogen and hydrophobic bonds. Generally, the stability of the former against environmental changes is higher than that of the latter. The latter could be useful when the release is switched by synergistic phenomena such as the phase transition of the wall membrane since the transitions are often reversible.
10.2.3 Evaluation of Rheological Properties of Microcapsules
Since most microcapsules for common use have a low rigidity, we often need to take account of only tension. The micropipette method is useful for determining the tension of a microcapsule wall membrane. In this method, a micromanipulator is operated with a micropipette and placed on a microcapsule to apply negative pressure, and the membrane tension is deduced from the observed length of the part withdrawn into the pipette, the radius of the capsule, and the curvature of the membrane. It was applied for estimating the membraneous tension of nylon microcapsules [20] and the deformability of erythrocytes [21]. The measurement of cell filterability using fabricated nickel mesh membranes is also useful for assessing cell deformability with high precision and has been used in studying the pathophysiology of various diseases [22]. When materials are released from microcapsules with a solid wall membrane owing to the breaking of the wall membrane by external pressure, the release features depend on microcapsule diameter and wall membrane thickness. Breaking stress increases with increasing microcapsule diameter at a constant membrane thickness, as well as with increasing membrane thickness at a constant microcapsule diameter [23].
The elastic modulus K of microcapsules in a suspension can be measured by a combination of ultrasonic velocity (V) and density (ρ) measurements. The typical frequency of ultrasonic waves used in laboratories is within 1–10 MHz, which corresponds to a wavelength range of 0.15–1.5 mm in aqueous solutions. Thus, the Rayleigh scattering approximation holds for microcapsules with a typical size of 1 μm, which is much smaller than the wavelength. Then, we can assume an additivity by volume to evaluate the adiabatic compressibility (B s = 1/K = 1/ρV 2) of the microcapsule membrane [24], which is the key parameter for the pressure sensitivity of microcapsules. As will be discussed in the next section using typical examples, the wall membrane of microcapsules is often in a swollen gel state, and the structure of the wall membrane could vary depending on the core-/wall-forming material weight ratio r [25]. Therefore, on the basis of this principle, the elastic modulus of each part of a microcapsule such as the swollen wall membrane and the inner core with a known volume fraction can be estimated [26]. The acoustic method using Brillouin scattering is used to measure microcapsule elasticity in the high frequency range. Relaxation dynamics in poly(lactide-co-glycolide) microcapsules with perfluorooctyl bromide was measured to be as high as 18 GHz [27].
When the inner core of microcapsules becomes nonvolatile on desiccation, the viscoelasticity of the microcapsules can be assessed by measuring the viscoelasticity of filter paper or a glass fiber blade on which microcapsule suspensions are spread [28].
Single-capsule measurement using the shell mechanics theory has been reviewed in several books [29]. The measurement of deformation of polyelectrolyte multilayer capsules has been demonstrated by this method [30]. The recent development of scanning microscopy now enables us to obtain the spatial distribution of viscoelasticity. Atomic force microscopy (AFM) developed originally for hard crystals [31] is also useful for soft hydrogels in liquid to determine the elasticity of microcapsules directly. The surface of microcapsules used as a bio-artificial pancreas was scanned, and a three-dimensional image was observed in a relatively early stage of development of AFM [32]. Recently, by combining AFM with fluorescence microscopy, the release of materials from the core of microcapsules with a polyelectrolyte multilayer membrane and plastic deformation was simultaneously measured [33].
Optical coherence microscopy (OCM) is a noninvasive noncontact in vivo imaging technique used to observe regions inside our body such as skin layers and the brain, using near-infrared light sources [34]. Using this technique, we can also obtain the elasticity distribution of a microcapsule surface. Optical coherence elastography (OCE) is one of the techniques of OCM and uses optical coherence tomography to measure the strain distribution by speckle tracking [35].
10.3 Relationship Among Preparation, Structure, Properties, and Function of Microcapsule
A large number of microencapsulation methods such as physical methods of air-suspension coating, coacervation-phase separation, centrifugal extrusion, and pan coating spray-drying and chemical methods of solvent evaporation, interfacial polymerization, and in situ polymerization have been developed. Even if limited to the use of only one of these methods, there will still be a variety of microcapsules with different structures and characteristics. Since the function of microcapsules and their mechanism depend on the structure and characteristics of microcapsules, the relationship among the preparation, structure, properties, and function of each type of microcapsule needs to be examined. Here, we consider typical conventional microcapsules prepared by interfacial polymerization in O/W or W/O emulsion: microcapsules with typical but very different characteristics, poly(urea-urethane) (PUU) microcapsules, and poly(L-lysine-alt-terephthalic acid) (PPL) microcapsules. A series of systematic studies from preparation to function have been performed on these microcapsules. A PUU wall membrane consists of a covalently cross-linked network with no charge. A PPL wall membrane consists of oligomers cross-linked by hydrogen bonds with a net charge depending on pH. In both phases, the reactants are low-molecular-weight compounds. When one of the reactants is composed of macromolecules, completely different phenomena occur, i.e., we can obtain microcapsule membranes with a unique anisotropic structure, since the conformation of macromolecules is modified on the contact surface, through which only low-molecular-weight compounds or ions except macromolecules diffuse, as shown in Fig. 10.2. This type of microencapsulation, i.e., dialysis-induced gelation on the surface, is also introduced.
Fig. 10.2
Illustration of microencapsulation by insolubilization
10.3.1 Microcapsules Prepared by Interfacial Polymerization
1.
Poly(urea-urethane) microcapsules
Microcapsules with poly(urea-urethane) (PUU) wall membranes are widely used as temperature-sensitive recording materials. The rate of thermal response is higher when the membrane thickness is smaller and the surface area is larger. Considering typical microcapsules with a diameter of 1 μm and a membrane thickness of several tens of nanometers, the rate of temperature change is estimated to be 150 K/ms. Thus, the temperature of the membrane could be changed instantaneously above the glass transition temperature of PUU from room temperature. This allows us to control the release of functional materials such as pigments and dyes from a microcapsule core by changing temperature, resulting in a change in the wall membrane permeation coefficient. Since PUU is biocompatible, microcapsules with a PUU wall membrane are also expected to be used as biomaterials such as drug delivery carriers [36].
They can be synthesized by interfacial polymerization in O/W emulsion, e.g., triisocyanate monomers in the organic phase react with immersing water at the interface of the oil phase and water phase to produce a microcapsule membrane. On the basis of the expected release rate of chemical agents such as pigments, an appropriate set of bulky stable organic solvents has often been used in industry.
Figure 10.3 shows cross-sectional views of microcapsules for systems with different volume fractions of the wall component and core component, as observed by scanning electron microscopy by the freeze-fracture method. The surface of the microcapsule membrane is clear for (a), and the membrane seems to be melted and very soft for (b) suggesting a swollen gel state. A noninvasive method of determining the structure of microcapsules in suspensions is a scattering method and a combination of a scattering method with other techniques [37]. Microcapsules with diameters in the order of 0.1 μm were investigated by the synchrotron small-angle X-ray scattering (SAXS) method and a combination of the SAXS and light scattering [38] methods, and those with diameters in the order of 1 μm and wall membrane thicknesses in the order of 0.1 μm were measured by a single-particle light scattering method [25a]. The swelling ratio of microcapsules consisting of a dioctyl phthalate core and an outer PUU membrane measured using single-particle light scattering and a freeze-fracture method in combination with electron microscopy shows that the swelling ratio of the wall membrane is as large as 1.2 [26b]. According to the theory of the swelling of a polymer network [39], the swelling ratio is related to the difference in the interaction parameter χ or the solubility parameter σ between the polymer and the solvent. According to the Fedors method [40], the σ values of a triisocyanate monomer and dioctyl phthalate are calculated to be 12.0 and 9.6 (cal/cm3)1/2, respectively. From the large difference in σ, the swelling ratio of 1.2, as determined by the light scattering method, seems to be too large. The microspheres prepared using only triisocyanate monomers did not swell so much in dioctyl phthalate. This apparent inconsistency might be related to the mechanism of interfacial polymerization and suggests that the presence of organic solvents affects the structure and physicochemical properties of microcapsule membranes. The resultant effects could be generalized to explain the mechanism of interfacial polymerization. On the other hand, such effects can be utilized to control the microcapsule structure and ultimately its function. Thus, it is of great importance to clarify the correlation of chemical cross-linking and molecular interactions between the wall-forming materials and the dispersing media in the mechanism of microencapsulation. It was also observed that the distribution of the core materials is not necessarily homogeneous. An elemental specific imaging technique in transmission electron microscopy showed that in microcapsules consisting of a PUU membrane and a triphenyl phosphate (TPP) core, TPP was found to be more concentrated in the vicinity of the outermost and innermost layers of the wall membrane [41], indicating the cross-linking density of the wall membrane also being inhomogeneous.
Fig. 10.3
(From Ref. [25a], Fig. 3) Cross-sectional views of PUU microcapsules with dioctyl phthalate (DOP) core prepared with different ratios of DOP and that of the wall-forming materials of triisocyanate monomer: (a) ϕ = 1.64 and (b) 7.81, as observed by scanning electron microscopy with the freeze-fracture method for system (a)
The glass transition temperature of a PUU microcapsule membrane depends on the core materials as a result of the plasticizing effect, when the solubility parameter of the core materials approaches that of the PUU network [25b]. A dynamic viscoelasticity study and scanning electron microscopy showed that PUU microcapsules containing tricresyl phosphate (TCP) did not form a core-shell structure but formed a swollen microsphere when the overall weight ratio of the core to the wall-forming material r < r critical = 0.5. The difference in the threshold ratio was reasonably explained by specific interactions such as the hydrogen bonding of the polymer network and the core materials through dipole moment and cohesive energy [25b]. With a core consisting of a compatible solvent, a homogeneous elastic membrane is formed by swelling, whereas with a core consisting of a less compatible solvent, the membrane becomes brittle and heterogeneous owing to phase separation. The rate of dye release from the microcapsules could be varied by more than 20-fold by varying the affinity between the core and the wall-forming materials [42].
The structural change of the wall membrane affects the mechanical properties such as the compressibility of microcapsules. An application of the ultrasonic method to suspensions of microcapsules consisting of a TCP core and an outer PUU membrane with a known volume fraction was performed, and the compressibility of the wall membrane was determined as a function of r [26].
Most of the release behaviors of microcapsules observed in experiments have been analyzed by kinetic theories based on Fick’s law with the assumption that the release rate is proportional to the concentration gradient of solutes, i.e., the ideal gas approximation [2, 43]. Microcapsules are, however, considered to be more complex systems; solutes such as staining materials could be distributed in the core and the membrane of microcapsules as well as in the dispersing medium with different fractions even in the equilibrium state. Furthermore, the compositions of each domain could change in the time course of the release. Thus, it is necessary to employ more strict treatment to understand the release mechanism and to characterize the release properties of such complex systems clearly. The theoretical results indicated that the release curve is expressed by the same exponential form as that for the theoretical one under ideal gas approximation but the time constant is related to the concentration derivatives of the chemical potential of solutes [44]. The difference should be clear when the concentration derivatives of the chemical potential vary significantly near the critical solution point of the core solution.
All of the above theoretical analyses are valid for a collection of microcapsules with a monodisperse size distribution. In real applications, however, the release rate is not precisely controlled because of the size distribution of synthetic microcapsules. Since relaxation time depends on the size of the microcapsules, “many relaxation times” may exist for a system of polydisperse microcapsules. On the basis of an analogy to relaxation phenomena with multiple characteristic times [45], theoretical arguments showed that the release curve can be expressed with a stretched exponential function with an effective time constant. Thus, the release curve for polydisperse microcapsules can be estimated from the size distribution and release behavior of monodisperse microcapsules, as shown in Fig. 10.4 [46, 47].
Fig. 10.4
(From Ref. [46], Figs. 3 and 4) (a) Diameter distribution curves for microcapsules consisting of PUU membrane and DOP core containing dye with narrow size distribution (open circle) and those with wide size distribution (closed circle) and (b) corresponding dye-release curves during exchange of medium from water to methanol. ρ and ρ eq denote dye concentration at each time and in the equilibrium state, respectively. The solid curve of (a) is that calculated from the dash-dotted line of (a) and the release profile of (b)
Since the PUU wall membrane has no charge, the stability of PUU microcapsule suspensions can be realized owing to the entropic effect of protective colloids such as poly(vinyl alcohol-vinyl acetate). For higher stability, hydrophilic polymers are often attached covalently to the membrane. Poly(ethylene glycol)s (PEGs) are one of the suitable candidates for this purpose [48].
2.
Poly(L-lysine-alt-terephthalic acid) microcapsule
Poly(L-lysine-alt-terephthalic acid) (PPL) microcapsules are prepared by copolymerizing L-lysine and terephthalic acid on the surface of aqueous droplets in W/O emulsion and replacing the oil phase with an aqueous solution, systematic studies of which were performed by Kondo, Makino, and Ohshima [49–52]. Since the component oligomers of PPL microcapsule membranes have amino groups and carboxylic acid groups at their terminals and many carboxylic acid groups along their skeleton [50], the polymer molecules interact with each other by hydrogen bonds, resulting in a three-dimensional polymer network to produce a microcapsule wall membrane. The electric charge of PPL microcapsule membranes changes in response to the pH of the medium, since the amino groups are protonated and carboxylic acid groups dissociate in neutral and alkaline media, while those near the isoelectric point between pH 2.3 and 2.6 charges are canceled out. The results of a single-particle light scattering method showed that PPL microcapsules with a thin membrane (σ ~ 0.2–1.2 μm) containing a buffer solution core undergo a volume phase transition as a function of medium pH [51]. Then, both the outer diameter of the microcapsules and the thickness of the wall membrane change, as shown in Fig. 10.5, resulting in the volume of PPL microcapsule membranes exhibiting marked changes in a narrow pH range between 5.0 and 6.0; thus, the size of PPL microcapsules is maximum at approximately pH 8 and minimum at approximately pH 4.5, where Makino et al. suggested the formation of a certain ion complex structure [52].
Fig. 10.5
(From Ref. [51], Fig. 7) Outer radius r o (closed circle) and membrane thickness δ (open circle) determined by single-particle light scattering of PPL microcapsules. Error bars denote the standard deviations of δ. The standard deviations of r o are smaller than the size of the circles
Gels that exhibit an immediate response, such as swelling or shrinking, to changes in the environment are materials of great potential and application [53]. For example, with such gels, chemo-mechanical systems, responsive artificial actuators, and carriers of drug delivery systems are expected to be developed further [54]. The characteristic time of swelling or shrinking of bulk gels is proportional to the square of the equilibrium size of the gel [55]. The behavior intrinsic to the gel could also be observed in a thin gel. Dynamic processes in the swelling and shrinking of a hydrogel membrane of PPL microcapsules are very different from those of bulk gels for the following reasons. (1) The microcapsule core serves as a proton reservoir in the swelling process (Fig. 10.6a) and as a hydroxyl anion reservoir in the shrinking process. This unique characteristic affects the transfer of the charge from the outside to the inside of the membrane. (2) The microcapsule membrane is extremely thin. Thus, collective diffusion is not time consuming, and the ion concentration gradient in the membrane is negligible. Theoretical studies showed that the dynamic processes of the swelling and shrinking of a PPL microcapsule hydrogel membrane can be characterized by two time constants related to the thin wall membrane and the large proton reservoir of the core [56]. Experimental swelling and shrinking processes are well explained by theoretical equations, as shown in Fig. 10.6b.
