Addition of Polysaccharides Influences Colloidal Interactions during

Langmuir 2008, 24, 8923-8928

8923

Addition of Polysaccharides Influences Colloidal Interactions during Latex Film Formation along with Film Morphology and Permeability Lisa Adersjo¨,† Johan Hja¨rtstam,‡ Bjo¨rn Bergenståhl,† and Lars Nilsson*,† DiVision of Food Technology, Faculty of Engineering LTH, Lund UniVersity, P.O. Box 124, 22100 Lund, Sweden, and AstraZeneca R&D Mo¨lndal, 431 83 Mo¨lndal, Sweden ReceiVed March 6, 2008. ReVised Manuscript ReceiVed April 8, 2008 In this paper, the effect of two polysaccharides (chitosan and dextran) on latex film morphology and porosity is investigated with atomic force microscopy, and the water permeability of the films is examined as well. Furthermore, latex films formed with mixtures of dextran and poly(ethylene glycol), PEG, are investigated. The results show that latex films without added polymers have the most homogeneous and dense morphology. In films containing dextran the highest degree of flocculation is observed, while these films do not show the highest water permeability. The highest permeability is observed in films containing chitosan and film porosity and permeability correlate positively to increasing chitosan concentration. The permeability of the latex films containing dextran and PEG accelerates with time. Since addition of these polymers to latex suspensions give rise to different morphologies and film permeabilities, this approach has promising abilities for control of film properties and, thus, has potential within controlled drug release.

Introduction The aggregation of colloidal particle dispersions and the underlying mechanisms for aggregation is of a great importance to a range of both fundamental and applied systems. Aggregation is typically caused by a reduction of repulsive forces between particles that is usually a result of changed composition and properties of the continuous phase and/or the surface of the particles. This involves an increase in ionic strength, change of solvent, increase in temperature, etc. If aggregation is desired, it can also be induced by the addition of soluble polymers. Added polymers that adsorb at the particle surface can give rise to both attractive and repulsive forces, depending on surface coverage.1 Hence, to obtain attraction, surface coverage should be incomplete, allowing for the formation of polymeric bridges between particles. Another way of introducing interparticle attraction is through the addition of polymers that are noninteracting with the particle surface, causing depletion attraction.2,3 A process that ultimately leads to aggregation is, naturally, if the continuous phase of the dispersion is evaporated. This occurs, for instance, in the formation of latex particle films, which are commonly created from latex suspensions upon evaporation of the continuous phase. The structure of films, formed from colloidal particle dispersions, depends on the ordering of particles during the evaporation step.4,5 In turn, the ordering of particles depends on the type of interaction between particles both in the initial suspensions as well as during the drying step. Typically, repulsion between particles results in ordered and dense films, while attraction results in fractal structures and porous films.4 Latex particles often have more or less charged surfaces that have a stabilizing effect on the suspension. Thus, a common way of * Corresponding author. Phone: +46 46 2228303. Fax: +46 46 2224622. E-mail: [email protected]. † Lund University. ‡ AstraZeneca R&D Mo¨lndal. (1) Guzey, D.; McClements, D. J. AdV. Colloid Interface Sci. 2006, 128, 227– 248. (2) Asakura, S.; Oosawa, F. J. Chem. Phys. 1954, 22, 1255–1256. (3) Scheutjens, J.; Fleer, G. J. AdV. Colloid Interface Sci. 1982, 16, 361–380. (4) Joanicot, M.; Wong, K.; Maquet, J.; Chevalier, Y.; Pichot, C.; Graillat, C.; Lindner, P.; Rios, L.; Cabane, B. Prog. Colloid Polym. Sci. 1990, 81, 175–183. (5) Chevalier, Y.; Pichot, C.; Graillat, C.; Joanicot, M.; Wong, K.; Maquet, J.; Lindner, P.; Cabane, B. Colloid Polym. Sci. 1992, 270, 806–821.

manipulating the interaction in such systems is by the addition of salt. Another way of manipulating the particle-particle interaction in a latex suspension would be to add an oppositely charged polymer, and for an anionic surface, this could, for instance, be chitosan, which interacts strongly with negatively charged surfaces.6,7 The effect of added chitosan to dispersions of negatively charged particles can result in both repulsion and attraction, depending on surface coverage,1 as mentioned above. Depletion attraction could also be an interesting way of generating attraction during film drying. An example of a suitable polymer is dextran, which is a noncharged polysaccharide known to induce depletion attraction between surfaces.8 Polymeric coatings, such as films formed from latex suspensions, have been used in a number of industrial areas for many years. One specific field of application involves their use as delayed or sustained release films in tablets as well as friction reduction during swallowing. The release characteristics of coated formulations are strongly dependent on the properties of the film, such as the water permeability and mechanical strength. The most common coatings are produced from organic solutions. Cellulose derivatives dissolved in ethanol-based solutions are used as coating materials in different extended-release formulations.9 However, for environmental reasons it is necessary, in the near future, to utilize water-based film-forming systems as, for instance, from aqueous latex suspensions. Atomic force microscopy (AFM) is a technique that has previously have shown to be a valuable tool for studying and characterizing latex films.10–12 With AFM it is possible to obtain high-resolution three-dimensional images without any pretreatment of the samples and, when using AFM in tapping mode, damage of film surfaces will be minimal and the films can easily be re-examined.10,12 (6) Fa¨ldt, P.; Bergenståhl, B.; Claesson, P. M. Colloids Surf. A 1993, 71, 187–195. (7) Mun, S.; Decker, E. A.; McClements, D. J. Langmuir 2005, 21, 6228– 6234. (8) Perez, E.; Proust, J. E. J. Phys. Lett. 1985, 46, L79-L84. (9) Rowe, R. C. Int. J. Pharm. 1986, 29, 37–41. (10) Wang, Y.; Juhue, D.; Winnik, M. A.; Leung, O. M.; Goh, M. C. Langmuir 1992, 8, 760–762. (11) Butt, H. J.; Kuropka, R.; Christensen, B. Colloid Polym. Sci. 1994, 272, 1218–1223. (12) Budhlall, B. M.; Shaffer, O. L.; Sudol, E. D.; Dimonie, V. L.; El-Aasser, M. S. Langmuir 2003, 19, 9968–9972.

10.1021/la800706m CCC: $40.75  2008 American Chemical Society Published on Web 07/23/2008

8924 Langmuir, Vol. 24, No. 16, 2008

Adersjo¨ et al.

The aim of the present paper is to investigate and understand how the morphology and permeability of latex films can be influenced and controlled by the addition of dissolved polymers. The methodology of the paper evolves around AFM and water permeability measurements of free casted films.

Materials and Methods The latex suspension, Kollicoat SR 30 D, was obtained from BASF AG, Ludwigshafen, Germany. The dispersion consists of about 27 wt % poly(vinyl acetate) and the stabilizers poly(vinyl pyrrolidone) (2.7 wt %) and sodium dodecyl sulfate (0.3 wt %). The minimum film formation temperature of the pure dispersion is 18 °C and the density is 1.045-1.065 g/cm3.13 The ζ-potential of the latex particles was determined with a Zetasizer 4 (Malvern Instruments, Malvern, UK) to be -2.0 mV at pH 3 and -4.6 mV at pH 6. Chitosan (Kitonor) was obtained from Norwegian Chitosan, Gardermoen, Norway. The degree of deacetylation is 90%, according to the supplier. Dextran 10 000 was obtained from Sigma-Aldrich Chemical Co., St Louis, MO. PEG 600 was obtained from Merck Schuchardt, Hohenbrunn, Germany. A schematic phase diagram was constructed in order to determine the phase boundary in the phase-segregating system dextran/PEG 600/water. Samples with different compositions were prepared and left under quiescent conditions to evaluate if separation occurred. For most of the cases, it was possible to determine if phase separation had occurred with only ocular observation. The samples that were close to the phase boundary in some cases needed to be observed by light microscope to be able to determine if the mixture contained one or two phases. Cover glasses for light microscopy (dimensions 24 × 32 × 0.13 mm) where used as substrates for latex films after they had been washed with chromosulfuric acid. The latex films were made from diluted latex dispersions. A small amount of the dispersion (∼75 µL) was deposited on a cover glass. The drop was spread out with an edge of another acid-washed glass to distribute the dispersion as thin and even as possible. The films were allowed to dry in an exicator at 22-23 °C for at least 2 h. After drying, the films were weighed and the mass per substrate area was estimated. Films from latex dispersions with added chitosan, dextran, or dextran/PEG were made in the same way as pure latex films. The films were analyzed using AFM (Nano Scope IIIa, Digital Instruments, Santa Barbara CA) with an aluminum-coated silicon tip in tapping mode. For each film, scans were done at four or five different positions. All images were modified by a second-order flattening. The average film thickness was estimated using

δ)

Γ F

(1)

where δ is the average film thickness (m), Γ is the mass per substrate surface area (mg/m2), and F is the density of the latex (kg/m3). Free films were prepared by spraying latex dispersion on a rotating cylinder, coated with poly(tetrafluoro ethylene) (PTFE), at room temperature.14 The dispersion was pumped to the spray nozzle, where pressurized air scattered the suspension as droplets. The distance between the nozzle and the rotating cylinder was approximately 13 cm. Drying of the film was sped up partly by having warm water inside the cylinder during spraying and partly by having a heating gun blowing warm air at the film. After drying, the film was peeled off the cylinder. The free films used for the water permeability measurements need to be thicker than films used for the AFM investigations in order to be able to peel them off the substrate. The initial latex concentration used for free films is 15 wt % in all cases. Chitosan was added at either 1 or 1.5 wt %. Dextran was added at 2.5 wt %, while dextran/PEG 600 was added at 2.5 wt %, respectively. The water permeability of the free films was studied using a diffusion (13) Technical Information for Kollicoat SR 30 D; www.pharma-solutions. basf.com, BASF AG, accessed 2006. (14) Allen, D. J.; Demarco, J. D.; Kwan, K. C. J. Pharm. Sci. 1972, 61, 106&.

chamber with two cells.15 The thickness of each film was measured at five different positions, with a digital micrometer, in order to obtain an average value. For each film two diffusion measurements were run simultaneously in two diffusion chambers. A segment of the free film was placed between the two cells. Fifteen milliliters of deionized water was simultaneously added to both cells to avoid the creation of a pressure over the film. Two paddles were used to stir the water in the cells and a jacket of water with the temperature 37 °C covered the chamber to maintain a constant temperature during the measurement. After some minutes, 10 µL of tritium-labeled water was added to the donor cell. At specified time intervals (∼30 min), samples of 500 µL were taken from the receiver cell and replaced by the same amount of deionized water. The samples were weighed and mixed with scintillation fluid before being analyzed in a Wallace Win Spectral 1414 liquid scintillation counter (PerkinElmer, Boston, MA). From the obtained values of disintegrations/ minute, the amount of labeled water that has diffused over the film can be calculated, from which the flow is obtained. The permeability was calculated from

Dperm )

Qh A

(2)

where Dperm is the permeability (m2/s), Q is the volumetric flow (m3/s), h is the film thickness (m), and A is the area of the cross section between the two cells were the film segment is positioned (m2). In our measurements, A ) 4.78 × 10-5 m2. The value used for calculation of the permeability is an average flow during the 3 h of the diffusion experiment, except for films containing dextran/ PEG 600.

Results The morphology of films formed from pure aqueous dispersions and with added chitosan, dextran or mixtures of dextran/PEG is studied using AFM. Films from pure aqueous latex dispersions without added polymer is used as reference films in all cases. In Figure 1, AFM micrographs of such film surfaces formed from 0.3 and 15 wt % latex dispersions are shown. From Figure 1 it is obvious that the dispersion is polydisperse (or possibly has a bimodal size distribution). The smaller particles have a diameter of approximately 100 nm, determined from section analysis; see Figure 1a. This average size is confirmed by dynamic light scattering (results not shown). Hardly any deformation or coalescence of the particles can be observed in the AFM images. The reference films are densely packed, although some small holes and cracks can be seen in some of the AFM images. No specific packing order can be observed. Also the films are not perfectly even and smooth, as height differences can be seen in the micrographs, where dark areas correspond to low parts and light areas to higher parts. The average film thickness is estimated using eq 1. For a film formed from 0.3 wt % latex, the average thickness is ∼200 nm and a film formed from 15 wt % latex has an average thickness of ∼2 µm. Chitosan is added to the latex dispersion in order to switch the particle surface charge from weakly anionic to strongly cationic. The films show an increasing porosity with increasing chitosan concentration in the interval 0.02-0.05 wt % (Figure 2). It should be noted that the films made from latex/chitosan are all prepared from dispersions with a chitosan concentration above the assumed amount required for full coverage of the latex particles surface (∼1 mg/m2) and, hence, bridging can be avoided. The dispersions are stable and homogeneous before drying when observed both macroscopically as well as with light microscopy. Films formed from latex dispersions with added dextran (Figure 3) are flocculated and show little or no dependence on dextran (15) Hja¨rtstam, J.; Hjertberg, T. J. Appl. Polym. Sci. 1999, 74, 2056–2062.

Polysaccharides Influence Latex Film Formation

Langmuir, Vol. 24, No. 16, 2008 8925

Figure 1. AFM micrographs of reference films. (a) The film made from a dispersions with 0.3 wt % latex (section analysis included). The scan size is 3 µm and the z-range 75 nm. (b) The film made from a latex with the same concentration as in part a but with a scan size of 15 µm and a z-range of 120 nm. (c) The film made from a 15 wt % latex dispersion. In the image, the scan size is 15 µm and the z-range 60 nm.

concentration in the investigated concentration range. Most likely the aggregation is caused by the increase in dextran concentration during film formation and, hence, the critical flocculation concentration (CF) cannot be avoided. To avoid this unintended phase separation a low molar mass PEG (M ) 600 g/mol) is incorporated in the system. The schematic phase diagram is presented in Figure 4. As can be seen from the phase diagram, it is not possible to avoid phase separation during film formation by addition of low amounts of PEG 600. The estimated phase boundary in Figure 4 agrees with the ternary phase diagram established by Olsson et al.16 for water, dextran (Mw ) 10 000 g/mol) and PEG 4000 or PEG 100 000. On the basis of the phase diagram established by Olsson et al., samples with high dextran concentrations are not investigated in the present paper. Films are formed from mixtures of 0.3 wt % latex, 0.05 wt % dextran, and 0.05 wt % PEG 600 or from 0.3 wt % latex, 0.1 wt % dextran, and 0.1 wt % PEG 600. The AFM micrographs of the obtained films (Figure 5a,b) show very interesting morphology. The characteristic packing pattern of the latex particles is not seen, but another pattern, which can be described as spongelike, is visible. The films at low dextran and PEG 600 concentrations show domains that are approximately 1–3 µm (Figure 5a). With (16) Olsson, M.; Joabsson, F.; Piculell, L. Langmuir 2005, 21, 1560–1567.

Figure 2. AFM micrographs of latex films formed from a 0.3 wt % latex dispersion with chitosan. The chitosan concentration and z-range, respectively, in each micrograph are (a) 0.02 wt % and 80 nm, (b) 0.03 wt % and 110 nm, (c) 0.05 wt % and 180 nm, and (d) 0.1 wt % and 210 nm. The scan size for all micrographs is 15 µm.

higher concentrations of dextran and PEG, somewhat larger domains are observed (approximately 5–10 µm) (Figure 5b). Free films where formed by spraying a latex suspension onto a rotating cylinder, as described in the Materials and Methods section of this paper. Latex films without added polymers are brittle and slightly rough. Brittle films are also obtained when the films are formed in the presence of chitosan or dextran. However, when mixtures of dextran and PEG 600 are used, the resulting film is rather elastic, showing that PEG functions as a plasticizer. The results from the water permeability measurements are presented in Figures 6 and 7. All films, except the latex/dextran/ PEG film, show linearity between the amount of labeled water that has diffused through the film and time. The film with latex/

8926 Langmuir, Vol. 24, No. 16, 2008

Adersjo¨ et al.

Figure 5. AFM images of films made from (a) 0.3 wt % latex, 0.05 wt % dextran, and 0.05 wt % PEG 600 and (b) 0.3 wt % latex, 0.1 wt % dextran, and 0.1 wt % PEG 600. Scan sizes in both micrographs are 15 µm and the z-ranges are 80 and 100 nm in parts a and b, respectively.

Figure 3. AFM micrographs of latex films formed from a 0.3 wt % latex dispersion with an increased amount of dextran. The dextran concentration and z-range, respectively, in each micrograph are (a) 0.05 wt % and 200 nm, (b) 0.1 wt % and 220 nm, and (c) 0.3 wt % and 200 nm. The scan size for all micrographs is 15 µm.

Figure 6. Diffused water volume vs time for free films consisting of pure latex (2), latex film with 2.5 wt % dextran (O), and latex film with 2.5 wt % dextran and 2.5 wt % PEG 600 (9).

to a relative permeability of about 10-1. The aggregated film formed with dextran is, on the other hand, less permeable. The permeability is only about 3 times higher than that of the reference film. The aggregated latex/dextran/PEG initially has a low relative permeability, similar to that of the reference film, which increases with time.

Discussion Figure 4. Schematic phase diagram for dextran/PEG 600/water, where CF denotes the critical flocculation concentration.

dextran/PEG shows a very different permeability profile (Figure 6) with a permeability that accelerates with time. This shows that components of the film are dissolved during the permeability measurement. Since the diffusion still seems to increase after the last measurement point, the dissolution can be considered incomplete at that time. Hence, two values of the flow through the film are used for calculating the permeability. Pure latex films display a permeability in the range of 7× 10-12 m2/s, which corresponds to a relative permeability of about 3 × 10-3. The aggregated film formed with the addition of chitosan gives about a 30-fold increase in the permeability, corresponding

Both the AFM micrographs (Figure 1) and the results from the permeability measurements show that latex films without additives are the densest and have the lowest permeability. The addition of chitosan, dextran, or dextran/PEG 600 influences both the morphology of the latex films as well as their water permeability. The films showed low permeability for latex without added polymers, medium permeability for latex with dextran, and high permeability for latex with chitosan. The addition of dextran/PEG 600 resulted in an initially less permeable film than with dextran as the sole additive. Furthermore, the permeability of the latex film with dextran/PEG 600 is accelerating with time. A system of monodisperse latex particles with significant repulsive interaction may pack with a crystalline order, leading

Polysaccharides Influence Latex Film Formation

Langmuir, Vol. 24, No. 16, 2008 8927

Figure 8. Schematic illustration of how chitosan (a) and dextran (b) can be distributed in and around latex particle clusters in the films.

Figure 7. The relative permeability for free latex films (15 wt % latex) with added chitosan (1 and 1.5 wt %), dextran (2.5 wt %), and dextran/ PEG 600 (2.5 wt %, respectively). *Relative permeability at time < 30 min. **Relative permeability at 165 min. The self-diffusion coefficient of water, DH2O, at 25 °C was used (2.3 × 10-9 m2/s).

to very dense films.4,17 However, the presence of a polydispersity disturbs the packing order and leads to more glasslike disordered films.18,19 A loss of order may also be obtained when repulsion between particles is insufficient.20 Furthermore, kinetic factors may also influence the ordering of particles. If packing occurs rapidly, insufficient time will be available to arrange particles in a dense and ordered way. In our system, a random glasslike packing is obtained, as seen in Figure 1. The polydispersity, the low surface charge, and a relatively short drying time may all contribute to this. Attractive forces between particles, resulting in flocculation, give films that are porous. In this study this is manifested by films containing dextran or chitosan. As dextran cannot be expected to adsorb to any large extent at the latex particle surface,21 it gives rise to depletion flocculation when the critical flocculation concentration is passed during film drying. A definite explanation to the aggregation in the latex films with chitosan is difficult to obtain. One possibility is that only partial chitosan coverage of the particle surfaces is reached. Thus, flocculation could be caused by bridging of chitosan between particles. However, as the amount of chitosan added should be sufficient to obtain full coverage of the surface, this explanation is less likely but cannot be completely ruled out. As the surface charge of the latex particles is very low, it is possible that complete adsorption of chitosan molecules does not occur. The mixing procedure when preparing samples most certainly affects the adsorption kinetics. If the adsorption rate is low, bridging can take place in the early phase of mixing. However, the adsorption rate increases with increasing concentration of the adsorbing species and, thus, the observed increased flocculation with increased chitosan concentration (Figure 2) is not supported by this explanation. The AFM micrographs of the films with the two highest chitosan concentrations have some similarities (Figure 2c,d). Both films are more like the films formed in the presence of dextran than the films with lower chitosan concentration. Hence, a possible explanation is that a depletion mechanism lies behind (17) Lee, J. M.; Kim, J. H.; Ho, C. C.; Cheong, I. W. Polymer 2007, 48, 4804–4813. (18) McRae, R.; Haymet, A. D. J. J. Chem. Phys. 1988, 88, 1114–1125. (19) Zohrehvand, S.; Cai, R.; Reuvers, B.; Nijenhuis, K. T.; de.Boer, A. P. J. Colloid Interface Sci. 2005, 284, 120–128. (20) Ro¨dner, S.; Bergstro¨m, L. J. Colloid Interface Sci. 2003, 265, 29–35. (21) Fournier, C.; Leonard, M.; Lecoqleonard, I.; Dellacherie, E. Langmuir 1995, 11, 2344–2347.

this aggregation. Depletion aggregation in polyelectrolytestabilized colloidal dispersions can arise when the particle surface is covered with polyelectrolyte and the amount of free unadsorbed polyelectrolyte reaches some critical level.1 Depletion aggregation has, for instance, been observed in SDS/chitosan-stabilized oilin-water emulsions.7 Another explanation is that the observed flocculation is related to capillary-induced phase separation (CIPS). CIPS is a manifestation of capillary condensation and results in the formation of a new phase between close-by particles, which in turn lowers the interfacial free energy of the system.22 For mixtures of dissolved polymers and colloidal particles, CIPS occurs when the polymer solution is close to phase separation and the polymer has affinity for the surface of the particles.23 The free energy density of the formed capillary phase is higher than in the surrounding reservoir phase; hence, the system strives to minimize the volume of the capillary phase. This, in turn, results in a long-range attractive force. The range of the attractive force far exceeds the range of both depletion and bridging forces.24 Thus, during the drying stage of the films containing chitosan, the phase separation boundary for the chitosan solution will be passed at some given point, possibly giving rise to capillaryinduced flocculation. The high permeability observed with chitosan agrees with a model suggesting that the pores in the film are free from chitosan. When comparing the AFM micrographs of the films with dextran (Figure 3) and the films with chitosan (Figure 2), the former case seems more porous than the latter. However, the permeability results show the opposite results; i.e., films containing chitosan have a higher permeability. The most likely explanation to this is that areas observed as pores in the dextran films in reality contain a thin layer of dextran molecules, which have been excluded from the interparticle regions during the depletion flocculation. This is illustrated schematically in Figure 8. The dextran layer, thus, would fill the pores between the latex aggregates and would thereby decrease the permeability of the film. The presence of PEG seems to have an effect both on the film morphology and the ductility of the film. The films containing PEG are elastic and much stronger than the other films in this study. This is attributed to the strong plasticizing properties of PEG. In the AFM micrograph in Figure 5, the morphology seen is completely different from the other films studied in this paper and has characteristics that could be described as spongelike. It is obvious from the morphology that the latex is distributed in one of the segregated phases formed during the drying of the dextran/PEG solution. In the AFM images in Figure 5, the darker (22) Evans, D. F.; Wennerstro¨m, H., The Colloidal Domain: Where Physics, Chemistry, Biology and Technology Meet, 2nd ed.; Wiley-VCH: New York, 1999. (23) Olsson, M.; Linse, P.; Piculell, L. Langmuir 2004, 20, 1611–1619. (24) Wennerstro¨m, H.; Thuresson, K.; Linse, P.; Freyssingeas, E. Langmuir 1998, 14, 5664–5666.

8928 Langmuir, Vol. 24, No. 16, 2008

Adersjo¨ et al.

areas would then correspond to the regions with latex-free solidified polymer and the brightest areas would correspond to the fused latex network. It can be expected that the solidified polymer network dissolves gradually when the film is rehydrated and this is shown as an increasing permeability with time. Such gradual permeability increase is not observed in latex/dextran films and is, thus, likely to be caused by the presence of PEG in the pores. In order to compare the different films in this study, it is useful to compare their apparent porosity. The highest particle volume fraction that can be obtained for close packed hard spheres, i e., hexagonal packing, is 0.74. The remaining 26% can, thus, be interpreted as a porosity value. Assuming that the porosity of the films is isotropic, that good connectivity exists between pores, that the pores are much larger than the diffusing molecules, and that tortousity is low, the porosity of the films in this paper can be estimated from

P ) DH2Oε

(3)

where P is the permeability (m2/s), DH2O the self-diffusion coefficient of water (m2/s), and ε the porosity value. The estimation shows that all films in this study have a lower porosity than 0.26. A possible explanation to this could be that deformation of latex particles has occurred. Another possibility is that some coalescence between particles has occurred, although coalescence was not observed in the AFM micrographs. However, as only the upper layer of the film can be observed with AFM, particles

beneath the surface may still be deformed. It should also be noted that the free films were made at a slightly elevated temperature since the rotating cylinder that the films were sprayed on contained hot water in order to speed up drying. Hence, free films could be more coalesced than films cast for AFM experiments.

Conclusion In this paper, we study the influence that added chitosan, dextran, and dextran/PEG have on the morphology and water permeability of latex films. The results show that it is possible to obtain films with different morphologies and porosities with addition of the above-mentioned polymers. Hence, these additives can be used to obtain the desired film permeabilities and, thus, be applied in controlled drug release, where films with higher relative permeabilities are expected to give a faster release. In this sense, the latex/dextran/PEG system is particularly interesting, as it initially has a very low permeability that increases rapidly after about 1 h. Acknowledgment. Carmen Carla Quiroga (Division of Food Technology, Lund University, and Food and Natural Products Center, San Simon University, Cochabamba, Bolivia) is acknowledged for her help with the atomic force microscope. John Janiak (Physical Chemistry 1, Lund University) is acknowledged for help with dynamic light scattering. LA800706M