Anal. Chem. 1998, 70, 111-116
Microspectroscopic Analyses of Dye Distribution Characteristics in Single Microcapsules Haeng-Boo Kim, Shuri Yoshida, Atsushi Miura, and Noboru Kitamura*
Division of Chemistry, Graduate School of Science, Hokkaido University, Sapporo 060, Japan
The distribution characteristics of a dye in single melamine resin wall microcapsules containing a dye/toluene solution are studied by a laser trapping-absorption microspectroscopy technique. In the case of Disperse Orange 13 as a dye, the molar absorptivity determined from the slope of a dye absorbance-capsule diameter (d) plot agrees well with that observed in a homogeneous toluene solution, indicating that the dye is solubilized homogeneously in the inner toluene solution of the capsule. For tetraphenylporphyrins (MTPP: ZnIITPP, CoIITPP, or H2TPP), however, the d dependence of absorbance shows a sigmoidal curve. Since absorbance of a ZnTPP/toluene droplet before polymerization of the melamine resin increases linearly with an increase in the droplet diameter, the sigmoidal d dependence of the absorbance is concluded to originate from distribution of the dye into the resin wall. Detailed analyses of the data indicate that the partitioning ratio of the dye between the toluene and resin phases is dependent on the capsule diameter. Size-dependent distribution of MTPP in single microcapsules is discussed on the basis of a proposed model and the d dependence of the dye absorbance, and it is concluded that the distribution of MTPP to the melamine resin wall is facilitated for larger microcapsules. Microcapsules possess a unique three-dimensional structure, in which a solution containing an arbitrary solute is surrounded by a spherical polymeric resin wall. Owing to this unique structure and a capability for arbitrary release of solute molecules to the outer phase, capsules have been widely used as foods, drugs, dry copies, coatings, and so forth.1 Furthermore, capsules have been studied as a model of artificial blood,2 and its applications to liquid-liquid extraction have been also explored.3 Commonly, microcapsules are prepared in water-in-oil or oilin-water emulsions, where polymerization of monomers proceeds around water or oil microdroplets containing a solute. Due to such a preparation method, it is believed that a solute is entrapped homogeneously in the inner solution of a capsule and its concentration is the same between the capsules. This is the (1) Lim, F. Ed. Biomedical Applications of Microencapsulation; CRC Press: Boca Raton, FL, 1984. (b) Kondo, A. Microcapsule Process and Technology; Marcel Dekker: New York, 1979. (2) (a) Muramatsu, N.; Kondo, T. J. Biomed. Mater. Res. 1980, 14, 211-224. (b) Kidokoro, M.; Ohshima, H.; Kondo, T. J. Microencapsulation 1991, 8, 63-70. (3) Watarai, H.; Hatakeyama, S. Anal. Sci. 1991, 7, 487-489. S0003-2700(97)00743-9 CCC: $14.00 Published on Web 01/01/1998
© 1997 American Chemical Society
fundamental basis for actual use of capsules: chemical and physical properties are the same between capsules. Nevertheless, this is not necessarily warranted, as reported by Koshioka et al.4 They demonstrated that the efficiency and dynamics of pyrene excimer formation in melamine resin/toluene microcapsules were different between the capsules, as studied by time- and spaceresolved fluorescence spectroscopy, and suggested that the results were due to a variation of pyrene concentration between the capsules. Although their results are quite interesting and involve aspects that are important to understanding chemical and physical properties of microcapsules, the use of pyrene excimer formation as a probe is an indirect method to evaluate a solute concentration. Clearly, more explicit studies are necessary to determine a solute concentration in individual microcapsules. Also, the role of the polymeric resin wall in the chemical and physical characteristics of capsules is worth being investigated. In this paper, we report direct analysis of solute distribution characteristics in individual melamine resin microcapsules by a laser trapping-absorption microspectroscopy technique.5 Laser trapping is a potential means to choose one microcapsule among others and to suppress its thermal Brownian motion in solution.6 Thus, precise and reproducible single-microparticle measurements are made possible by combining laser trapping with absorption spectroscopy.7,8 Furthermore, since the laser trapping spectroscopy technique enables one to measure individual microdroplets before polymerization of the melamine resin wall, the role of the polymeric wall in the dye distribution characteristics would be also studied. Generally, solute distribution characteristics in a single microparticle cannot be examined by absorption spectroscopy. According to our model and theoretical calculations, however, a study on the particle diameter dependence of UV-visible absorbance (Abs) of a solute can provide information about the solute’s distribution characteristics.9 In the case of microcapsules, namely, when the capsule wall is thin compared to the capsule diameter (d) and the solute is solubilized homogeneously in the inner (4) Koshioka, M.; Misawa, H.; Sasaki, K.; Kitamura, N.; Masuhara, H. J. Phys. Chem. 1992, 96, 2909-2914. (5) Preliminary results have been reported: Kim, H.-B.; Yoshida, S.; Miura, A.; Kitamura, N. Chem. Lett. 1996, 923-924. (6) Masuhara, H., De Schryver, F. C., Kitamura, N., Tamai, N., Eds. MicrochemistrysSpectroscopy and Chemistry in Small Domains; North Holland: Amsterdam, 1994. (7) Kitamura, N.; Hayashi, M.; Kim, H.-B.; Nakatani, K. Anal. Sci. 1996, 12, 49-54. (8) Kitamura, N.; Nakatani, K.; Kim, H.-B. Pure Appl. Chem. 1995, 67, 79-86. (9) Kim, H.-B.; Yoshida, S.; Kitamura, N. Anal. Chem. 1998, 70, 51-57 (in this issue).
Analytical Chemistry, Vol. 70, No. 1, January 1, 1998 111
Scheme 1. Preparation of Dye/Toluene Containing melamine Resin Microcapsules
solution, the d dependence of Abs should obey Lambert-Beer’s law, with d correspondeding to the optical path length. If a solute is distributed exclusively to the resin wall, on the other hand, the slope of an Abs vs d plot becomes one-third of that expected for a homogeneous solute distribution in the inner solution. On the basis of a study on the d dependence of Abs, therefore, we can determine both concentration and distribution characteristics of a solute in single microcapsules. In the following, we demonstrate a unique application of the laser trapping-absorption microspectroscopy technique to studying chemical and physical characteristics of single microcapsules in solution. EXPERIMENTAL SECTION Preparation of Dye/Toluene-Containing Microcapsules. Melamine resin wall microcapsules containing a dye/toluene solution were prepared according to the method of Koshioka et al.4 with some modifications (Scheme 1). An aqueous solution (60 mL) of melamine (2.4 × 10-3 mol) and formaldehyde (6.9 × 10-3 mol) was mixed with acacia (natural rubber acts as a surfactant, 0.5 g in 10 mL of water), and the pH of the solution was adjusted to 4 -5 with phosphoric acid. Into the solution was added a toluene solution (10 mL) containing an appropriate amount of a dye (concentration, C0), and the mixture was stirred vigorously (12 000 rpm) for 10 min at room temperature to prepare an oil-in-water emulsion. Polymerization of the melamine resin was performed by heating the emulsion at 65 °C for 2 h under mechanical stirring (200 rpm) in the presence of ammonium sulfate as an initiator (3.0 × 10-4 mol). After cooling, the reaction mixture was poured into a sufficient amount of toluene-saturated water as a sample. The diameter of the capsules ranged from 8 to 30 µm, and the capsules were stable for several days. All of the experiments were performed with freshly prepared samples at ambient temperature. Chemicals. In this study, we chose an organic azo dye (Disperse Orange 13, DO13) and tetraphenylporphyrins as solutes 112
Analytical Chemistry, Vol. 70, No. 1, January 1, 1998
to encapsulate in the polymeric resin wall, owing to their high molar absorptivities in the visible region. The use of these dyes is very suitable for accurate and precise absorption measurements of single microcapsules. DO13 (Aldrich, GR grade) was used as supplied. Zinc, cobalt, and metal-free tetraphenylporphyrins (ZnTPP, CoTPP, and H2TPP, respectively; Wako Pure Chemicals, GR grade) were used without further purification. Spectroscopic grade toluene (Wako Pure Chemicals) was used as supplied. Pure water was obtained by deionization and distillation (Advantec, GSR-200). Laser Trapping-Absorption Microspectroscopy. An experimental setup for laser trapping-absorption spectroscopy of single microparticles in solution has been reported elsewhere.7,8 A 1064-nm laser beam from a CW Nd:YAG laser (Spectron, SL902T) was used as a trapping light source. The laser beam was introduced to an optical microscope (Nikon, Optiphoto 2) and focused (∼1-µm spot) by an objective of the microscope (×100, NA ) 1.30). The laser power was adjusted to 1 W throughout the experiments. Under the present conditions, the actual laser power irradiated to a microcapsule was several tens of milliwatts, as reported previously.10 For absorption spectroscopy, a Xe light beam (Hamamatsu Photonics, L2274) was introduced to the microscope coaxially with the laser beam and irradiated onto an optically trapped microcapsule. The probe beam diameter and quality are very important in order to perform accurate absorption microspectroscopy. In the present experiments, a paraxial ray of the microscope objective was used as a quasi-parallel probe beam, and its diameter was adjusted to ∼1 µm. The beam passed through the capsule (intensity, I) and condenser lens, and a pinhole was reflected by a half mirror set under the microscope stage and led to a multichannel photodetector (Princeton Instruments, ICCD-576E/C or IRY-512G with an Oriel Multispec 275 or Jovin-Yvon 320 polychromator) via an optical fiber to record an absorption spectrum. The incident light intensity of the Xe beam, I0, was determined under the same optical conditions without a microcapsule.11 Further details on the optical requirements have been reported in separate publications.7-9 RESULTS AND DISCUSSION Laser Trapping-Absorption Spectroscopy of Single DO13/ Toluene-Containing Microcapsules. Laser trapping of single microcapsules containing a dye/toluene solution was achieved by irradiating the capsule with a focused 1064-nm laser beam as reported previously.12 With the thermal Brownian motion of the capsule being suppressed, we performed absorption spectroscopy on individual microcapsules dispersed in toluene-saturated water. As a typical example, an absorption spectrum of a DO13/toluene capsule (diameter (d) ) 32 µm) is shown in Figure 1a. Although the spectrum observed for the capsule was slightly broader than that determined in a homogeneous toluene solution, the peak (10) Misawa, H.; Koshioka, M.; Sasaki, K.; Kitamura, N.; Masuhara, H. J. Appl. Phys. 1991, 70, 3829-3836. (11) With decreasing the capsule size, chromatic aberration becomes important, and this leads to more or less distortion of a spectrum, as reported previously.7,9 However, since the absorption spectral band shape of the dye was independent of the capsule size with the d range of 8-30 µm, such an effect is neglected in the present experiments. (12) Misawa, H.; Kitamura, N.; Masuhara, H. J. Am. Chem. Soc. 1991, 113, 78597863.
Figure 1. Absorption spectrum of DO13 in a single toluenecontaining melamine resin microcapsule (capsule diameter (d) ) 32 µm, (a)) and a plot of the capsule diameter dependence of the absorbance at 427 nm (b). Individual microcapsules were laser trapped by a 1064-nm laser beam.
Figure 2. Absorption spectra of ZnTPP in a single toluenecontaining melamine resin microcapsule (d ) 19 µm, solid curve, 1.9 × 10-3 M) and in a homogeneous toluene solution (broken curve, observed in a 10-mm-optical path length cuvette, 1.8 × 10-6 M).
wavelength (λmax ) 427 nm) agreed very well with that in toluene, irrespective of d. If d corresponds to the optical path length (l), then the absorbance of the dye (Abs) should increase linearly with an increase in d. Actually, we obtained a good linear relationship between Abs and d, as shown in Figure 1b. We determined the molar absorptivity of DO13 in the capsule as ) 2.7 × 104 M-1 cm-1 (427 nm), under the assumption that the concentration of the dye was the same as that in the mother toluene solution (C0 ) 1.0 × 10-2 M). However, the value was larger than that determined for a toluene solution (2.1 × 104 M-1cm -1) by ∼30%. The linear relationship in Figure 1b proves homogeneous distribution of DO13 in the inner toluene solution, since the capsule wall is thin (∼100 nm)2 compared with d (i.e., l ∼ d). Also, there is no probable reason for an enhancement of the value in the capsules. It is noteworthy that an analogous enhancement of has been suggested for pyrene/toluenecontaining microcapsules.4 As discussed later again, we suppose that the dye is concentrated during the preparation of the capsules, which involves heating at 65 °C for 2 h (Scheme 1).
Figure 3. Capsule diameter dependencies of absorbance of MTPP: (a) ZnTPP (at 424 nm, C0 ) 8.7 × 10-4 M), (b) CoTPP (at 415 nm, C0 ) 1.7 × 10-3 M), (c) H2TPP (at 420 nm, C0 ) 7.8 × 10-4 M). The broken and dotted lines represent d dependencies of the absorbance predicted from E ) 1 and 0 (eq 2), respectively. The solid curves are the best-fit curves by eqs 2 and 3. See also the main text.
Diameter Dependence of Absorbance for MTPP/TolueneContaining Microcapsules. A spectrum of a single ZnTPP/ toluene-containing capsule (d ) 19 µm) dispersed in toluenesaturated water is shown in Figure 2, together with that observed in a homogeneous toluene solution (1.9 × 10-3 M). The Soret and Q bands of ZnTPP in the capsule can be seen at maximum wavelengths of 424 and 550 nm, respectively. Although the spectrum of the capsule shifts slightly to the shorter wavelength, the spectra agreed very well with each other. Similarly, the maximum wavelength of the Soret band of CoTPP or H2TPP in the capsules was determined to be 415 or 420 nm, respectively. The spectral band shape of each MTPP in the capsule was independent of d, similar to the results for DO13. The most interesting result is a d dependence of Abs. As the data for ZnTPP, CoTPP, and H2TPP are summarized in Figure 3 (a, b, and c, respectively), the d dependence of Abs did not coincide with that expected for a homogeneous dye distribution in the capsule (broken line in the figure).13,14 Analogous tendencies were observed at several MTPP concentrations (discussed later). One possible reason for this is distribution of MTPP in the surrounding water phase. However, MTPP was scarcely distributed to water, as confirmed by separate extraction experiments. Furthermore, the figures indicate that, although Abs tends to increase with an increase in d, that extrapolated to d ) 0 does not cross the original point of the plot. Namely, the deviation between the observed and predicted (broken line) Abs at a given d increases with an increase in d. Clearly, the results cannot be explained on the basis of distribution of MTPP in the water phase. It is concluded that MTPP is not distributed in the inner toluene (13) The values used for the calculations were 5.4 × 105 (424 nm), 2.5 × 105 (415 nm), and 4.8 × 105 (420 nm) M-1 cm-1 for ZnTPP, CoTPP, and H2TPP, respectively. (14) Although the dye solution is condensed during the synthetic procedures as discussed in the main text, its efficiency in each capsule cannot be known. Therefore, a simulation of the dye absorbance-d plot was performed with the MTPP concentration in the mother solution, C0.
Analytical Chemistry, Vol. 70, No. 1, January 1, 1998
113
2 Abs ) dC0 1 - E 3
(
Figure 4. Droplet diameter dependence of absorbance of ZnTPP (424 nm) in single toluene microdroplets. The sample was the same as the emulsion before the polymerization. [ZnTPP] ) 9.7 × 10-4 M. The solid line represents the d dependence of the absorbance calculated by the Lambert-Beer law; Abs ) [ZnTPP]d, with ) 5.4 × 105 M-1 cm-1.
solution of the capsule alone, which is in marked contrast to the results for the DO13-containing capsules in Figure 1. The above discussion suggests that MTPP should distribute to the melamine resin capsule wall. In order to confirm this, we conducted absorption spectrum measurements on single ZnTPP/ toluene microdroplets. The sample was essentially the same with the oil-in-water emulsion before the polymerization (Scheme 1). As shown in Figure 4, Abs increased linearly with d, and the value ((4.9 ( 0.2) × 105 M-1 cm-1) determined from the slope of the plot and [ZnTPP] ) 1.0 × 10-3 M agreed with that for a homogeneous toluene solution (5.4 × 105 M-1 cm-1) within experimental error. The results prove that the melamine resin wall plays an essential role in the peculiar d dependencies of Abs in Figure 3. Furthermore, the agreement of the values obtained in the two experiments supports the conclusion that evaporation of the toluene solution during the capsule preparation is the main reason for the larger value in the capsules compared with that predicted from C0. When MTPP distributes to both toluene and resin phases, Abs determined under a microscope is given as in eq 1, as reported previously,9,15 where E is the partitioning ratio of MTPP between
{
Abs ) dC0
}
RE 1-E + 2 (1 - R) 1 - (1 - R)3
(1)
the two phases, defined as the ratio of the mole number of MTPP in the resin phase (Ns) to the whole capsule (N0): E ) Ns/N0. R is θ/r, where θ and r are the thickness of the capsule wall and the radius of the capsule (d/2), respectively. It is important to note that eq 1 indicates that Abs is dependent on both E and R. The thickness of the melamine resin wall is supposed to be ∼100 nm,2 so that θ is small compared to r (5-15 µm): R < 0.02. Therefore, eq 1 can be reduced to eq 2. This equation demon(15) For this equation to be applied, the radius of a probe beam (h) must be small compared with the radius of a capsule (r): h/r < 0.2. In the present experiments, the diameter of the probe beam was set to ∼1 µm, while that of the capsules studied ranged between d ) 8 and 25 µm. Therefore, this condition of h H2TPP > CoTPP. We suppose that the distribution efficiency is determined by molecular interaction between MTPP and the melamine resin, since the above sequence agrees very well with that of the electron-donating ability of MTPP: oxidation potential is 0.78, 1.02, or 1.10 V for Zn-, H2-, or CoTPP, respectively.17 The melamine resin is made of a syn-triazine structure known to be an electron-deficient heterocycle,18 so that an electron donor-acceptor-type interaction between MTPP and the resin will play an important role in determining the distribution efficiency of the dye from the toluene solution to the resin phase. Since DO13 is a very weak electron donor,19 an interaction with syn-triazine or melamine resin is not expected. The homogeneous distribution of DO13 in the inner toluene solution will be thus a reasonable consequence. We conclude that the chemical properties of the present dye/toluenecontaining melamine-resin microcapsules are determined during the synthetic procedures and by the chemical nature of the dye. CONCLUSIONS So far, chemical and physical properties of microcapsules have been discussed on the basis of the experimental results determined for a large number of capsules by particle-unresolved methods. Therefore, the results obtained by such studies are an ensemble average of those for the capsules with different properties. As discussed in the present report, chemical properties are not necessarily the same between the capsules, and they are shown to vary with the nature of the a solute, solute concentration, and capsule size. For both basic sciences and applications of microcapsules, an elucidation of chemical and physical properties as a function of the capsule size is quite important. It is noteworthy that, although confocal fluorescence microspectro Analytical Chemistry, Vol. 70, No. 1, January 1, 1998
115
scopy can monitor the three-dimensional structure of a microparticle, samples to be applied should be fluorescent.20 In this respect, we emphasize that a study on a particle diameter dependence of UV-visible absorbance is very important in order to elucidate inhomogeneous structures of nonfluorescent or weakly fluorescent microparticles. Besides capsules, the same will be true for various microparticles, such as droplets, polymer beads, biocells, and so forth. Application of the laser trapping microspectroscopy technique has high potential for such studies, and the work along that line is now in progress in this laboratory.
116
Analytical Chemistry, Vol. 70, No. 1, January 1, 1998
ACKNOWLEDGMENT N.K. is grateful for a Grant-in-Aid from the Ministry of Education, Science, Sports and Culture (08404051) for partial support of the research. Received for review July 10, 1997. Accepted October 6, 1997.X AC970743B X
Abstract published in Advance ACS Abstracts, November 15, 1997.
