Langmuir 2002, 18, 2873-2879
2873
Polymerization of Styrene in DODAB Vesicles: A Small-Angle Neutron Scattering Study Martin Jung,† Brian H. Robinson,‡ David C. Steytler,*,‡ Anton L. German,† and Richard K. Heenan§ Department of Polymer Chemistry, Eindhoven University of Technology, P.O. Box 513, 5600 MB Eindhoven, The Netherlands, School of Chemical Sciences, University of East Anglia, Norwich, Norfolk, NR4 7TJ, United Kingdom, and ISIS Facility, Rutherford Appleton Laboratory, Chilton, Oxon OX11 0QX, United Kingdom Received September 11, 2001. In Final Form: December 21, 2001 The polymerization of styrene, located in the bilayer of dioctadecyldimethylammonium bromide (DODAB) vesicles, gives rise to phase separation between the growing polymer and the bilayer. The result is a small (20-30 nm) bead of polymer located in the bilayer of each vesicle giving them a “parachute-like” appearance. Small-angle neutron scattering (SANS) has been used to monitor the evolution of the polymer morphology in situ, on the time scale of polymerization, and thus to characterize all intermediate morphologies. Analysis of the scattering profiles for the case of both photo and thermal initiation reveals that a phase separation sets in immediately after the onset of polymerization involving a fast nucleation interval followed by a slower growth phase. Above the phase transition temperature, the data support an even distribution of monomer within the vesicle prior to initiation. Growth of a single, oblate ellipsoid polymer particle then ensues with the asymmetry of the particle decreasing with increasing extent of polymerization. Neither the reaction temperature nor the mode of initiation (photo or thermal) was found to have a significant effect on this underlying mechanism of particle growth. Conversely, below the phase transition temperature SANS supports an uneven monomer distribution within the bilayer resulting in multiple nucleation and growth of more than one particle per vesicle.
Introduction Polymerization processes in vesicle bilayers have attracted much interest in the last two decades. The multidisciplinary research in this area has contributed to new insights not only in colloid and polymer chemistry but also in biophysics and material science.1-7 Early efforts were directed toward the stabilization of the vesicle bilayer through polymerization of the lipids making up the bilayer.1-4 More recently, attempts were made to solubilize standard monomers in vesicle membranes and to subsequently polymerize them inside the bilayer to increase the rigidity of the membrane. The concept was to use the vesicle as a template with a view to copying the vesicle structure as a polymeric material.8,9 Several research groups10-14 have claimed that hollow polymer particles * Corresponding author. † Eindhoven University of Technology. ‡ University of East Anglia. § Rutherford Appleton Laboratory. (1) Armitage, B. A.; Bennet, D. E.; Lamparski, H. G.; O’Brien, D. F. Adv. Polym. Sci. 1996, 126, 53-83. (2) O’Brien, D. Trends Polym. Sci. 1994, 2, 183-188. (3) Singh, A.; Schnur, J. M. In Phospholipids Handbook; Cevc, G., Ed.; Marcel Dekker: New York, 1993; Chapter 7. (4) Bader, H.; Dorn, K.; Hupfer, B.; Ringsdorf, H. Adv. Polym. Sci. 1985, 64, 1-62. (5) Ringsdorf, H.; Schlarb, B.; Venzmer, J. Angew. Chem. 1988, 100, 117-162. (6) Fendler, J. Membrane Mimetic Chemistry; Wiley: New York, 1983. (7) Fuhrhop, J. H.; Ko¨nig, J. Membranes and Molecular Assemblies: A Synkinetic Approach; Royal Society of Chemistry: Cambridge, 1994; Chapter 4. (8) Murtagh, J.; Thomas, J. K. Faraday Discuss. Chem. Soc. 1986, 81, 127-136. (9) Meier, W. Curr. Opin. Colloid Interface Sci. 1999, 4, 6-14. (10) Kurja, J.; Nolte, R. J. M.; Maxwell, I. A.; German A. L. Polymer 1993, 34, 2045-2049. (11) Poulain, N.; Nakache, E.; Pina, A.; Levesque, G. J. Polym. Sci., Part A: Polym. Chem. 1996, 34, 729-737. (12) Morgan, J. D.; Johnson, C. A.; Kaler, E. W. Langmuir 1997, 13, 6447-6451.
can be synthesized through this route. In particular, a recent study of catanionic surfactants employing smallangle neutron scattering (SANS) and cryo-transmission electron microscopy (cryo-TEM) supports formation of such hollow polymeric spherical structures.15 In contrast, our own investigations on polymerization in dodecyltrimethylammonium bromide (DTAB) vesicles led us to the conclusion that polymerization in vesicles causes phase separation of the polymer in the bilayer to form a small particle.16,17 An alternative approach has been to alter the solubilization characteristics of the monomer/polymer within the bilayer by using fluorinated lipids. In this way,18 it was shown that polymerization of isodecyl acrylate within perfluoroalkylated phosphatidyl choline (F-PC) vesicles results in a the desired “hollow shell” structure whereas the same reaction carried out in egg PC vesicles gives rise to the phase-separated morphology. It appeared that the phase separation phenomenon is very general and occurs for many polymer/amphiphile combinations irrespective of process parameters such as temperature, molecular weight of the polymer, or method of initiation.19,20 For the particular case of styrene polymerization in dioctadecyldimethylammonium bromide (DODAB) (13) Hotz, J.; Meier, W. Adv. Mater. 1998, 10, 1387-1390. (14) Hotz, J.; Meier, W. Langmuir 1998, 14, 1031-1036. (15) McKelvey, C. A.; Kaler, E. W.; Zasadzinski, J. A.; Coldren, B.; Jung, H.-T. Langmuir 2000, 16, 8285-8290. (16) Jung, M.; Hubert, D. H. W.; Bomans, P. H. H.; Frederik, P. M.; Meuldijk, J.; van Herk, A. M.; Fischer, H.; German, A. L. Langmuir 1997, 13, 6877-6880. (17) Hubert, D. H. W.; Cirkel, P. A.; Jung, M.; Koper, G. J. M.; Meuldijk, J.; German, A. L. Langmuir 1999, 15, 8849-8855. (18) Krafft, M. P.; Schieldknecht, L.; Marie, M.; Pascal, M.; Giulieri, F.; Schmutz, M.; Poulain, N.; Nakache, E. Langmuir 2001, 17, 28722877. (19) Jung, M.; Hubert, D. H. W.; van Veldhoven, E.; Frederik, P. M.; van Herk, A. M.; German, A. L. Langmuir 2000, 16, 3165-3174. (20) Jung, M.; Hubert, D. H. W.; Bomans, P. H. H.; Frederik, P. M.; van Herk, A. M.; German, A. L. Adv. Mater. 2000, 12, 210-213.
10.1021/la011419l CCC: $22.00 © 2002 American Chemical Society Published on Web 02/20/2002
2874
Langmuir, Vol. 18, No. 7, 2002
Figure 1. Schematic representation of two possible pathways of phase separation leading from a monomer-laden vesicle (top left) to parachute morphologies (top right and bottom left).
vesicles, we have found that the polymerization results in a parachute-like vesicle-polymer structure (so-called because of their appearance when visualized by cryoTEM). Up to now, the pathway and the mechanism of this intriguing parachute formation have not yet been established. Therefore, this study aims to elucidate the mechanism of phase separation and the involved time scales by an in situ study of the polymerization. Small-angle neutron scattering is a powerful method for this investigation. Using deuterated styrene in an otherwise hydrogenated system (h-DODAB in H2O), it is possible to focus the SANS measurements purely on the evolving polymer in the course of polymerization. The influence of the polymerization temperature and the type of initiation, thermal or photochemical, has been examined. Theoretical Considerations on the Mechanisms of Phase Separation: Monomer Diffusion versus Polymer Diffusion The starting point for polymerization of styrene in DODAB vesicles may be considered as a vesicle bilayer in which the monomer is uniformly distributed. This is a reasonable assumption for most cases21 as will be shown later. Theoretically, one could imagine at least two extreme scenarios that lead to the fully phase-separated parachute morphology (Figure 1): (i) (Steps 1-3) In the beginning, initiation occurs at one particular site in the vesicle bilayer. The growing polymer chain then attracts more monomer and growing oligomers, as they are probably more soluble therein than in the aliphatic vesicle bilayer. As a consequence, a polymer/ monomer microenvironment develops where propagation preferentially takes place due to an increased local monomer concentration. Fast migration of monomers and oligomers in the surfactant matrix21 (D ∼ 2 × 10-7 cm2/s) would allow a comparatively fast reorganization of monomer and oligomeric species22 and could support the polymerization in one nucleus, simultaneously inducing a depletion of monomer in the rest of the bilayer. This mechanism will be called the monomer diffusion mechanism. (ii) (Steps 4-5) The second mechanism assumes that polymerization starts throughout the vesicle bilayer and polymer chains are thus equally distributed over the (21) Jung, M.; Hubert, D. H. W.; van Veldhoven, E.; Frederik, P. M.; Blandamer, M. J.; Briggs, B.; Visser, A. J. W. G.; van Herk, A. M.; German, A. L. Langmuir 1999, 16, 968-979. (22) We take the circumference of the vesicle (π × 160 nm) as a maximum diffusional distance for a monomer or an oligomer to reach a growing nucleus. Considering a diffusion coefficient of D ) 2 × 10-11 m2/s for lateral diffusion, the involved time scale to circle around a vesicle then calculates as t ) (π × 160 nm)2/4D ) 1.3 × 10-2 s.
Jung et al.
bilayer on the time scale of polymerization. Subsequently, due to unfavorable polymer-surfactant interactions and restricted conformational freedom, the polymer chains gradually migrate to one point, a nucleus, where they coalesce into a particle, hence reducing surface contact area and surface free energy. This process could occur as a postpolymerization “ageing” effect or perhaps after achieving a certain critical degree of conversion during the polymerization process when the solubility in the monomer/bilayer environment declines below a limiting value. As this route depends on the polymer diffusion in the bilayer, we will call it the polymer diffusion mechanism. SANS provides a unique tool for discriminating between the two pathways by focusing on the morphological changes of the monomer/polymer during the polymerization and so identifying the structural change from an initial shell distribution to the phase-separated parachute morphology. Experimental Section Materials. DODAB (Acros, >99%) was used as received. Deuterated styrene (d8, Sigma, >98% D) was passed over an inhibitor removal column (hydroquinone removal column, Aldrich) prior to use. For thermal initiation, the water-soluble azoinitiator 2,2′-azobis(2-methylpropionamide)dihydrochloride (V50, Aldrich, 97%) was used as received. For photochemical initiation, the oil-soluble initiator 2,2′-dimethoxy-2-phenyl acetophenone (DMPA, Aldrich, 98%) was used without further purification. Super-Q (Millipore) water was used for vesicle preparation. Vesicle Preparation. Unilamellar vesicles were prepared by triple extrusion (under 7 bar argon pressure) of a dispersion of 10 mM aqueous DODAB through three stacked 200 nm polycarbonate filters (Millipore, hydrophilized PC filters) at 60 °C. Prior to extrusion, the dispersion was allowed to hydrate at 60 °C for 2 days. After extrusion, the vesicle dispersion was maintained at 60 °C for 1 day before slowly cooling to room temperature. The intensity-weighted z-averaged diameter of the vesicle population as determined by dynamic light scattering (Malvern 4700)21 was 159 nm. Polymerization. Prior to polymerization, oxygen was removed from the dispersion by repeated evacuation cycles followed by flushing with argon. For all experiments, the concentration of DODAB was 10 mM. Under stirring, d8-styrene was added to the vesicle solution at room temperature to obtain an overall monomer concentration of 20 mM. The dispersion was stirred for 2 days to solubilize the monomer. For thermal initiation, V50 was first dissolved in a small volume of water and then injected into the heated dispersion to give an initiator concentration of 5 mM. The polymerization temperature was 50 °C. For photochemical initiation, the initiator DMPA was first dissolved in the monomer at a concentration of 0.1 M. The resulting solution was then added to the vesicle dispersion at 25 °C and stirred for 2 days. For in situ measurements, solutions were prepared 3-5 days before the neutron measurements were made and then protected from light. Just before the photopolymerization experiment, the samples were filled in 1 mm thick Hellma circular quartz cells and mounted in a thermostated holder of the SANS beam line. A UV light source (HPR 125W, Philips) was located in a horizontal plane at 40 cm from the cell. The range of wavelengths of the lamp (P ) 5.1 W at λmax ) 366 nm) matched the UV-absorbing region of the photoinitiator. By moving the sample between a UV-illuminated zone and the neutron beam, intermittent irradiation of fixed duration could be provided and the resulting polymerization could be followed in situ by SANS. Quenched polymerization samples were also prepared in the Eindhoven laboratory before the SANS measurements. For these samples, photopolymerization was induced to a fixed level of conversion and was then interrupted. Further polymerization was prevented by addition of a small amount of a hydroquinone solution and subsequent protection from light.
Polymerization of Styrene in DODAB Vesicles Small-Angle Neutron Scattering. SANS measurements were performed on the LOQ instrument23 on the ISIS pulsed neutron source at the Rutherford-Appleton Laboratory, U.K. The samples were contained in 1 mm thick circular quartz cells (Hellma) and could be thermostated in a sample holder to better than 0.1 °C over the temperature range 10-60 °C by means of a Julabo FP 52 circulating bath. The measurements yield the absolute scattering intensity I(q) [cm-1] as a function of momentum transfer q ) 4π/λ sin(θ /2) [Å-1] with λ being the incident neutron wavelength (2.2 Å < λ 20 nm). Although the system is expected to be less well defined, the data could be more closely represented by an oblate
2876
Langmuir, Vol. 18, No. 7, 2002
Figure 3. SANS of d8-styrene in extruded DODAB vesicles at 25 °C: (a) log I(q)-log q representation showing a gradient close to -2 as expected for a “sheetlike” bilayer of a large vesicle; (b) form factor fit for a disk of infinite radius giving a bilayer thickness of 3.4 nm. [DODAB] ) 10 mM, [d8-styrene] ) 20 mM.
Figure 4. Best form factor fits to d8-styrene in extruded DODAB vesicles prior to polymerization at 10 °C. Dashed line ) infinite disk (5.0 nm thickness); solid line ) oblate ellipsoid (r1 ) r2 ) 30 nm, X ) 0.42). [DODAB] ) 10 mM, [d8-styrene] ) 20 mM. See text for details.
ellipsoid model, as used to model the phase-separated polymerized system (Figure 4). The parameters obtained for the oblate ellipsoid were with the equal half-axes (r1 ) r2) of the oblate at 30.0 nm and the short axis (r3 ) Xr1) defined by an accentricity factor X ) 0.42. 2. Quenched Photopolymerization at 25 °C. There are two obvious ways to follow the progress of the reaction. One is to quench the reaction after various time intervals, determine the conversion, and subsequently record the progress of the reaction by SANS. A second method is to illuminate the reaction for defined periods of time on the SANS instrument and record the SANS profile in situ.
Jung et al.
Figure 5. (a) SANS data for photopolymerization at 25 °C with samples quenched at different levels of conversion: experimental data (symbols), vertically displaced for clarity, and fitted form factor (lines) for oblate spheroids. (b) SANS from a similar single sample at 20% conversion illustrating the quality of fit over an extended q range using the ellipsoid model (r1 ) r2 ) 20.0 nm, X ) 0.57). [DODAB] ) 10 mM, [d8styrene] ) 20 mM.
The advantage of the first method is that one obtains morphological data as a function of conversion, whereas the second method allows on-line monitoring of the reaction, but without knowledge of the conversion. Both methods have been employed in this study. Photochemical initiation is expected to be a less complex process than thermal initiation. First, in the case of thermal initiation, there is the possibility of the nucleation occurring in the water phase external to the membrane. Second, there is evidence for the formation of bilamellar vesicles in the case of thermal initiation19 that may happen as a result of osmotic effects distorting the bilayer forming stomatocyte-type structures.31 Figure 5a shows the SANS data for a photochemical polymerization reaction quenched after fixed periods of continuous illumination following the start of the reaction. The progressive evolution in growth of the polymer beads is clearly indicated. Tests of fitting the data to various models showed that a form factor for oblate ellipsoids gave a good quality of fit. We were unable to reproduce the data as accurately using polydisperse sphere models represented by a variety of polydispersity functions. Furthermore, we feel a degree of confidence in adoption of the ellipsoid model since phase separation and particle growth take place within a bilayer structure that would naturally impose unsymmetrical constraints. However, we are aware of the ambiguity that exists between (31) Hubert, D. H. W. Surfactant Vesicles in Templating Approaches. Ph.D. Thesis, Eindhoven University of Technology, Eindhoven, The Netherlands, 1999; Chapter 9.
Polymerization of Styrene in DODAB Vesicles
Langmuir, Vol. 18, No. 7, 2002 2877
Table 1. Fitted Parameters Obtained from SANS Data for Different Polymerization Runs time [min]
conversion [%]
oblate fit analysis r1 ) r2 [nm] X req [nm]
5 15 45 60
Quenched Photopolymerization at 25 °C 7 20.0 0.31 22 20.0 0.56 51 23.0 0.54 78 25.0 0.56
13.5 16.5 18.7 20.6
1 2 3 4 7 17
On-Line Photopolymerization at 25 °C 20.0 0.27 22.0 0.36 21.0 0.46 22.0 0.49 24.5 0.57 26.0 0.57
13.0 15.7 16.2 17.3 20.3 21.5
1 5 27
On-Line Photopolymerization at 60 °C 20.0 0.41 23.5 0.56 26.0 0.54
14.9 19.4 21.2
Quenched, Thermally Induced Polymerization at 50 °C 18 21.0 0.47 16.3 51 24.6 0.58 20.5 94 29.5 0.58 24.6 On-Line Thermally Induced Polymerization at 50 °C 15 20.0 0.47 15.5 125 23.0 0.57 19.1
polydisperse sphere and ellipsoid form factors and acknowledge that a polydispersity function must exist that will fit the data equally well. The procedure adopted to fit the data was to progressively increase the value of the two equal half-axes of the oblate ellipsoid (r1 ) r2) and float the value of the short axis (r3 ) Xr1) until an acceptable fit was obtained. The fit results are shown in Figure 5a, and the derived parameters are presented in Table 1. For reference, we also give the oblate parameters translated into an equivalent sphere radius, req ) (r1r2r3)1/3. Support of the ellipsoid model adopted is given by the good quality of fit to a single SANS measurement made to lower Q values on the same system (Figure 5b). It appears that relatively flat oblates of 40 nm “diameter” (2r1) and 12 nm “height” (2Xr1) are formed within the early stage of polymerization which coincides favorably with the observations by cryo-TEM. Increasing conversion mainly leads to an expansion in height (nearly by a factor of 2) and only moderate growth in diameter, that is, approach to a more spherical structure. Interestingly, the calculation of the oblate volume as a function of conversion indicates that the particle volume (V ) (4/3)πr1r2r3) exceeds the estimated polymer volume as calculated by the conversion fraction of the final volume (Figure 6). We take this observation as a hint that the polymer could be swollen by monomer giving an additional volume contribution to the particle. With increasing conversion, this contribution becomes less significant. 3. On-Line Photopolymerization. Having established the morphology of the quenched vesicle-polymer hybrids, we performed SANS measurements using on-line polymerization. To assess the effect of the bilayer phase transition temperature, reactions were carried out at four temperatures (10, 25, 40, and 60 °C) reflecting conditions on each side of the phase transition temperature for the monomercontaining21 (∼27 °C) and polymer-containing vesicles19 (∼45 °C). The samples were illuminated by UV for burst periods of 1 min, and then the SANS was recorded with the sample removed from the UV light beam. SANS measurements were also made to check that the reaction made no progress within the “dark time” when the sample was not illuminated.
Figure 6. Particle volume as a function of conversion for the photopolymerization of d8-styrene in extruded DODAB vesicles at 25 °C. Polymerization was quenched at fixed levels of conversion. [DODAB] ) 10 mM, [d8-styrene] ) 20 mM.
A specimen set of results is shown in Figure 7a-d for polymerizations at 10, 25, 40, and 60 °C. At all temperatures, the reaction is essentially complete after a total illumination period of about 30 min. Inspection of the figures shows that the reaction proceeds through a steady increase in size of the evolving (poly)styrene particle. The oblate ellipsoid model was again found to adequately represent the scattering. Typically, the results show an increase in r1 ) r2 from about 20 to 25 nm while X increases from 0.31 to 0.56. We thus can picture a growth process from a rather flat oblate, that is, nearly a disk, which is progressively expanded in height to approach a more symmetric, spherical shape. There is no evidence for the spontaneous nucleation of a number of small particle sites in, or on, the bilayer throughout the process. A prolonged existence of a hollow polymer shell morphology cannot be reconciled with these observations, and the proposed monomer diffusion route thus appears unambiguously to be the more adequate model. The geometrical data obtained (cf. Table 1) allow calculation of the particle volume growth as a function of time which depends on two principal factors: (i) the volume growth rate of the polymer which is proportional to the rate of polymerization and (ii) the volume contribution by monomer that swells the polymer particle. Figure 8 compares the particle volume growth of polymerizations at 25 and 60 °C. As expected, the growth rate at 60 °C exceeds the growth rate at 25 °C due to a 4-fold increase in the propagation rate coefficient of styrene.32 However, the growth at 60 °C starts to slow at an earlier extent of the polymerization than at 25 °C. This could be explained by a decrease in monomer concentration at the site of polymerization at 60 °C arising from the higher solubility of styrene in water at higher temperatures. The main observation at both temperatures is the immediate occurrence of a polymer/monomer particle with dimensions that surpass the bilayer thickness, that is, with the particle significantly swelling the region of the bilayer in which it is hosted. We thus conclude that the mechanism involves fast nucleation followed by a slower growth phase. Changing the polymerization temperature does not change the morphological evolution, but it is found by cryo-TEM that the polymerization at 10 °C, well below the phase transition temperature of the monomer(32) Propagation rate coefficients, kp, for styrene are kp (25 °C) ) 85.9 L/(mol s) and kp (60 °C) ) 341 L/(mol s); see: Buback, M.; Gilbert, R. G.; Hutchinson, R. A.; Klumperman, B.; Kuchta, F. D.; Manders, B. G.; O’Driscoll, K. F.; Russell, G. T.; Schweer, J. Macromol. Chem. Phys. 1995, 196, 3267-3280.
2878
Langmuir, Vol. 18, No. 7, 2002
Jung et al.
Figure 8. Particle volume measured “on-line” as a function of time for the polymerization of d8-styrene in extruded DODAB vesicles at 25 and 60 °C. [DODAB] ) 10 mM, [d8-styrene] ) 20 mM.
Figure 9. Cryo-TEM micrograph after polymerization of styrene in DODAB vesicles at 10 °C showing multiple formation of polymer beads. The scale bar corresponds to 100 nm. [DODAB] ) 10 mM, [d8-styrene] ) 20 mM.
Figure 7. SANS data (line plus symbol) showing the effect of temperature on the “on-line” monitored photopolymerization of d8-styrene in extruded DODAB vesicles: (a) 25 °C, (b) 60 °C, (c) 10 °C, and (d) 40 °C. [DODAB] ) 10 mM, [d8-styrene] ) 20 mM.
containing vesicle system, leads eventually to smaller and less well-defined polymer particles (see Figure 9). In strong contrast to earlier observations at temperatures higher than 25 °C,16-20 we notice now that vesicles typically carry more than one polymer particle. Apparently, polymerization at temperatures below the phase transition induces multiple nucleation. This effect can primarily be ascribed to the drastically reduced lateral diffusion of monomeric and oligomeric species in the bilayer. Thus, if the time scale for diffusion of monomers and oligomers to one single polymer nucleus per vesicle becomes too long a new nucleus would be created within the bilayer. SANS also demonstrated the existence of monomer pockets, that is, small
phase-separated monomer islands, within the bilayer at 10 °C (Figure 4). These islands could help in creating multiple polymer nuclei. In accord with the observations of less distinct particles and formation of monomer islands, we experienced difficulty in obtaining satisfactory fits to the SANS data at 10 °C and do not report any fitted parameters for the particle growth. 4. Quenched, Thermally Induced Polymerizations. Results for quenched samples produced by thermally induced polymerization at 50 °C are shown in Figure 10. As for photochemical initiation discussed above, an oblate ellipsoid model was chosen to represent the data. Measurements were here made to lower Q values, and the quality of fits obtained over this extended Q range again lend support to the adoption of this model. From the fitted parameters (Table 1), the general pattern observed for photochemical initiation prevails, that is, rapid phase separation of a “flat” oblate entity followed by slow growth giving rise to decreased asymmetry. 5. On-Line Thermally Induced Polymerizations. Thermally induced polymerizations were also monitored online at 50 °C using the water-soluble azo-initiator V50. The scattering was recorded on-line using intervals of 5 min to obtain acceptable statistics, and the reported SANS therefore reflects an integration over all structures
Polymerization of Styrene in DODAB Vesicles
Langmuir, Vol. 18, No. 7, 2002 2879
Conclusions
Figure 10. Neutron scattering data at given time intervals for the quenched, thermally induced polymerization of d8styrene in extruded DODAB vesicles at 50 °C: experimental data (symbols), vertically displaced for clarity, and fitted form factor (lines) for oblate ellipsoids. [DODAB] ) 10 mM, [d8styrene] ) 20 mM.
Figure 11. Neutron scattering data recorded over 5 min time intervals during the on-line thermally induced polymerization of d8-styrene in extruded DODAB vesicles at 50 °C. [DODAB] ) 10 mM, [d8-styrene] ) 20 mM.
occurring within this time interval. Figure 11 shows a representative set of SANS data at different stages of the reaction for which the fitted parameters (Table 1) are in reasonable agreement with those obtained from thermally initiated, quenched samples. In essence, the data exhibit the same features as seen before for the photochemical initiation, although the mechanism of initiation is distinctly different. In the case of the water-soluble initiator, the initiation occurs through the water phase probably after entry of oligomeric radicals into the vesicle aggregate. However, the same morphological development of an expanding oblate geometry is extracted from the fits to the data. The markedly slower rate of particle volume growth compared to the photochemical polymerizations is readily explained by a smaller effective radical flux from the water phase.
The polymerization of styrene in DODAB vesicles is known to lead to phase separation between bilayer matrix and polymer particle for which two hypothetical routes of phase separation are possible (Figure 1). To differentiate between these, small-angle neutron scattering experiments can be used to monitor the evolution of the polymer morphology on the time scale of polymerization by contrasting the monomer/polymer against the surrounding reaction medium. The on-line experiments demonstrate that a polymer/ monomer nucleus is created immediately after initiation. The geometry of the nucleus could be described best by an oblate ellipsoid, possibly caused by the constraints of the bilayer medium in which it is born. Particle growth with proceeding conversion then occurs mainly by an extension in height until the particle attains a final geometry where the short axis amounts to about 60% of the long axes. This behavior was not influenced by the mode of initiation and the polymerization temperature, unless going to temperatures far below the phase transition where several polymer beads are formed on one vesicle. Altogether, the observations of the on-line experiments give compelling evidence of the suggested monomer diffusion model. A polymer shell morphology could not be detected at any moment of the reaction. This is entirely consistent with our earlier electron microscopic observations. In light of the results presented, it becomes easily understandable why even the application of cross-linking monomers cannot suppress the polymer-amphiphile phase separation: the interval of nucleation is too short and the further growth mechanism solely hinges on the diffusion of monomer and small oligomeric species to the nucleus. This suggests that attempts to slow polymer diffusion will leave the phase separation unaffected. It appears that the most promising approach to synthesize hollow polymer particles with this system is to tackle the problem of nucleation, by for instance the use of copolymerizable amphiphiles. Our current investigations are directed toward the exploration of such remedies against phase separation. Acknowledgment. We acknowledge the European Community for several TMR travel grants and the allocation of beam time at ISIS and HMI. This work was supported by The Netherlands Organization for Chemical Research (NWO/CW). We gratefully acknowledge assistance from Dr. Ernst Hoinkis (HMI) for SANS measurements and Paul Bomans and Dr. Peter Frederik, University of Maastricht, for the cryo-TEM analysis. LA011419L
