Polyelectrolyte multilayer hollow capsules studied by small-angle neutron scattering (SANS)

Abstract

Silica-templated polyelectrolyte (PE) hollow capsules with various template diameters and PE layer numbers were fabricated by using the layer-by-layer technique and subsequent core dissolution in hydrofluoric acid. The properties of the resulting freestanding polyelectrolyte multilayers (PEM) were characterized in an aqueous environment by means of SANS and compared with scanning force microscopy (SFM) data. The thickness of the capsule wall was found to be about 25% thicker and with correspondingly higher water content than the template-supported PE layers. The wall thickness increase as a result of core dissolution was anisotropic. The average single layer thickness of hollow capsules was independent of the surface curvature and decreased slightly with increasing PE layer number. SANS was used to determine whether the capsules were empty or contained the rest of the core. An annealing at 70 °C for 4 h induced capsule shrinking by about 20% at an ionic strength of 1 mol L⁻¹. Furthermore, the capsule wall upon annealing increased in thickness by about 38%. These changes corresponded to a wall densification of about 13%.

---

1. Introduction

Layer-by-layer adsorption of oppositely charged polyelectrolytes initially employed for flat surfaces¹ has nowadays become a well-established method for preparing colloid templated PEM.^2,3 Hollow capsules can be obtained from these templates by removal of sacrificial cores.^4,5 The capsules thus represent freestanding PE films.

A particular feature of PEM hollow capsules is that their composition, internal structure, thickness, density, porosity of PE layers, mechanical properties and capsule geometry can be tuned, depending on the core, the protocol of core dissolution and the coating materials.^6–22 Ionic strength and subsequent annealing were found to be quite important for controlling the permeability of capsule shells.^11,23–32 These novel capsule architectures have hence found large interest because of their possible use as drug-delivery systems with controlled release function, for diagnostic and sensing purposes.^33–42

Among the various possibilities colloidal silica spheres have been intensively used as templates after some chemical modification.^{38,39,43–51} Particles with a narrow size distribution are available in a wide range from less than 10² up to 10⁴ nm. The silicon dioxide core can be easily removed in hydrofluoric acid. A particular advantage of using silica templates is that during core dissolution in hydrofluoric acid the capsule wall experiences less mechanical and chemical stress in comparison with such templates as melamine-formaldehyde (MF) particles, polystyrene (PS) latex beads or erythrocytes.³⁸

In view of the envisaged applications of PEM capsules, information concerning capsule wall thickness, structure and density of the film is highly desirable. It would be interesting to compare these data for freestanding polyelectrolyte films such as in capsules with supported PE layer on the colloids and flat surfaces. Such knowledge would be essential for optimising the wall properties with regard to stability, thickness and permeability.

Supported planar PE multilayer films have been extensively studied by neutron reflectivity.^52–56 This technique is most appropriate to obtain data on thickness, structure and density of polyelectrolyte multilayer films. As a matter of fact, most of the current knowledge on thickness, density profile, water content and regularity of stratification of multilayers has been derived from neutron reflectivity studies. This powerful high resolution technique is not applicable to colloidal systems, such as PEM-coated colloids and capsules. Instead, SFM,^9,16 applied in the dry state on collapsed capsules; single-particle light scattering,³ a combination of flow cytometry, confocal laser scanning microscopy (CLSM) with quartz crystal microgravimetry (QCM)⁴⁹ have been used as techniques for the structural characterization and build up of PEM-covered colloids and capsules. However, for example, the thickness of PE layer derived from these techniques rests on a variety of assumptions, e.g. the water content of the capsule walls, their refractive index, assuming complete collapse etc. Quite recently, full-field transmission X-ray microscopy (TXM) has been applied for the determination of water content and thickness of the wall of hollow polyelectrolyte capsules⁵⁰ and polymeric microballoons.⁵⁷ This novel scanning imaging technique with a resolution currently between 10 and 40 nm is a perfect complement to the diffraction-based high resolution techniques, such as SANS.

Small-angle neutron scattering has proved to be a powerful method for the ultrastructural characterization of polymer-decorated colloids^58–65 and core–shell nanoparticles.^66–70 SANS has been applied to PEM-coated polystyrene latex⁵⁸ and more recently to PEM capsules templated on human erythrocytes.⁷¹ Beside measuring the wall thickness and the density profile with nanometre precision, SANS is capable of providing the water content of the wall and detecting remains of the core in the capsule interior.

We studied PEM capsule walls consisting of the 8, 12 and 16 layers of poly(styrene sulfonate (PSS) and poly(allylamine hydrochloride) (PAH). Sacrificial silica templates of 130 nm, 1.2 and 3 µm in diameter were used. The wall thickness and the content of the lumen of hollow capsules were characterized in an aqueous environment by means of SANS. The measurements were compared with SFM data. The layer thickness was determined as a function of layer number thus providing a detailed quantification of growth. We were further interested in how the capsule parameters depend on ionic strength and annealing.

2. Experimental

Hollow capsule preparation

Silica colloids with a diameter of 130 nm (small), according to data provided by the manufacturer, 1.2 µm (large), and 3 µm were used as templates. The layer-by-layer technique was applied for the alternating deposition of the PAH and PSS, starting with PAH. The adsorption of polyelectrolytes was conducted in 0.5 M NaCl solution for 15 min followed by washing three times in water. Silica colloids covered with 8, 12 and 16 layers were fabricated. The silica cores were dissolved in 0.1 mol L⁻¹hydrofluoric acid. The colloids were washed first with HF and then thoroughly rinsed with water until a neutral pH was reached. SANS contrast enhancement was achieved exchanging H₂O for D₂O solutions. Annealing was conducted incubating the capsules at 70 °C in 0.1 or 1 mol L⁻¹NaCl D₂O solution for 4 h.

Small-angle neutron scattering

SANS experiments were conducted at the Hahn-Meitner-Institute, Berlin, on the SANS instrument V4. The SANS instrument operates in the q-range of 0.04–3 nm⁻¹. q, the scattering vector, is given by q = 4πsin(θ/2)/λ, where θ is the scattering angle and λ is the neutron wavelength. Experimental data were collected with a sample-detector distance of 0.975, 4, and 15.85 m and a wavelength of 6.05 Å. The monochromator system is a mechanical selector, which gives a wavelength spread of Δλ/λ = 10%. Quartz cuvettes of 1 mm thickness were used. The raw scattering data were corrected for background from buffer, cuvettes, electronics and ambient noise. The remaining background was included in the curve fitting as a parameter. The scattering data were converted to the absolute scale by using a reduction procedure described elsewhere.⁷² All measurements were conducted in D₂O as solvent. Since the capsule concentration is not known with sufficient precision the SLD could not included as a fitting parameter, but was fixed at 4 × 10⁻⁶ Å⁻².⁵⁸

Scanning force microscopy (SFM)

SFM experiments were performed using the Molecular Force Probe (MFP) 3D instrument (Asylum Research, Santa Barbara, CA). The MFP was used in combination with an inverted optical microscope, Olympus IX 71 (3D), equipped with a 40× objective. Images of dry collapsed hollow capsules were recorded in contact mode using MSCT-AUHW cantilevers from Veeco Instruments (Woodbury, NY). The spring constant was determined for each cantilever separately using the thermal noise method. All measurements were carried out in air at room temperature. The layer thickness values of the dry capsules were deduced from the statistical analysis of height profiles of the capsule images.

3. Results and discussion

3.1 Scattering theory and data analysis

The model used for the description of scattering consists of the core (including a possible silica rest after core dissolution), the shell (PEM) and the bulk solution. Particle–particle interactions are negligible because a sufficiently small capsule concentration was used in all neutron experiments. Thus, only form factors provide the contributions to the coherent scattering intensity. These form factors describe the scattering due to the core (S_cc), the interference between the core and the shell (S_pc), and the polyelectrolyte shell itself (S_pp).

If the capsules were completely hollow, the capsules interior would then have the same scattering length density (n_c) as the bulk solution (n_s), and thus only the S_pp form factor contributes to the scattering intensity.

S_pp(q) = 2π(S/V)q⁻²|∫e^iqz〈ϕ(z)〉dz|² (1)

where S/V is the specific area, q is the scattering vector, 〈ϕ(z)〉 is the polymer volume fraction and z is the radial coordinate. If 〈ϕ(z)〉 is a step function (〈ϕ(z)〉 = ϕ_p for 0 ≤ z ≤ h and 〈ϕ(z)〉 = 0 for z > h), where h is the thickness of the polymer layer, eqn (1) reduces to:

S_pp(q) = 2π(S/V)q⁻⁴(2ϕ²)(1 − cos(qh)). (2)

After applying McLaurin series for the condition qh/2 ≪ 1, eqn (2) can be further reduced to:

q²S_pp(q) = 2π(S/V)ϕ_p²h²(1 − q²h²/12) (3)

with eqn (3) a rough estimation of the shell thickness, h, from the scattering intensity profile is possible.

Fig. 1 represents the scattering profiles of hollow capsules templated on small and large silica particles with eight deposited polyelectrolyte layers. For the sake of comparison, the scattering intensity provided by bare silica colloids is also shown. The position of the scattering minimum of the bare silica particles is consistent with a diameter of 102 nm, which is close to the size provided by the manufacturer obtained by means of dynamic light scattering. The slope of the scattering intensity curve from the bare particles is equal to −4 in the q-range corresponding to qR ≫ 1. The curve behaviour in this q-range satisfies Porod's law and proves the absence of an adsorbed layer on the surface of the colloids. An adsorption layer would give, according to eqn 1, a contribution in the scattering intensity scaling as q⁻².

Fig. 1 Small-angle neutron scattering intensity profiles corresponding to small (■), large (▲) silica-templated hollow capsules with 8 PE layers, and to small bare silica colloids (●). The solid lines denote the best fits of the experimental data including device resolution and polydispersity. SLD here and in subsequent fits was fixed at 4 × 10⁻⁶ Å⁻².⁵⁸

The scattering intensity profiles of the hollow capsules in Fig. 1 reveal a q⁻² law in the range of small q. In the particular case of small capsules the scattering minimum related to the size of the spherical colloids can be clearly observed in the q-range smaller than 0.2 nm⁻¹. The fitting procedure applied to the data in Fig. 1 yielded a diameter of the capsule lumen of 106 nm being consistent with the template size. As one might expect, fringes were not observed for the 1.2 µm templated hollow capsules. The large diameter of the templates would yield very small separations between scattering minima in q space. These minima are strongly smoothed by the finite device resolution and the capsule polydispersity. Thus, only for the small capsules there is a chance of determining the diameter of PEM capsules after core dissolution.

At q values larger than 0.2 nm⁻¹ at least one minimum occurrs as a result of the interference between the waves reflected from the two interfaces of the polymer shell (eqn (2)) can be distinguished. The position of this minimum is related to the layer thickness. This minimum is observed at the same q value independent on the size of the capsules indicating an independence of capsule wall thickness on curvature.

The wall thickness can be estimated either from the position of the minimum or by applying eqn (3) to the scattering function ln(q²I(q)) vs.q². Both estimations of the thickness were used as initial parameters for subsequent fitting.

A non-linear least square fitting protocol to a model function for polydisperse spherical particles with a core–shell structure was applied to the experimental data.^73–75 The polymer volume fraction profile was described by a step function. The polydispersity of the radius of the particles together with its coat was described by a Schulz–Zimm distribution and taken into account for fitting. The instrument resolution was also included by convoluting the scattering function with the device resolution function approximated by a Gaussian distribution. The variance of the resolution function was calculated from the wavelength distribution, finite collimation, aperture, and spatial resolution of the scattering instrument.^76,77 The solid lines in Fig. 1 denote the best fits of the experimental data. The fits reveal in the case of 8 polyelectrolyte layers a wall thickness of 180 ± 6 Å for the large, and 170 ± 7 Å for the small capsules, respectively. The empty capsules (n_c = n_s) model yielded the best fit to the experimental results.

SANS data can also reveal whether there will be remains of the core after dissolution and capsule fabrication. In this way it can be useful as an unambiguous quality check of the fabricated hollow capsules. Fig. 2 demonstrates an example of “bad” hollow capsules with a silica rest in their interior after core dissolution. The solid line indicates the core–shell model (n_c ≠ n_s) fitted to the experiment data, where all three scattering contributions are present. The dashed line refers only to the core contribution. At the I(q)vs.q presentation the scattering profile of the capsules clearly reveals a deviation from the slope “−2” at very small q values, contrary to what one would expect for empty capsules. Presenting the scattering data as ln(q²I(q)) vs.q in Fig. 2 gives the possibility to observe most clearly the influence of eventually present core scattering as a deviation from linear behaviour (eqn (3)) at q² < 0.02 nm⁻². In the particular case shown in Fig. 2 the scattering length density of the capsule interior obtained by fitting was about 5.86 × 10⁻⁶ Å⁻². Considering the core SLD as superposition of the contributions from deuterated water and the silica rest, and taking into account the SLD value of dry silica the volume fraction of silica rest after core dissolution in the capsule interior can be calculated. For example, the sample presented in Fig. 2 reveals incomplete core dissolution, as 18% of the silica is still found in the capsule lumen. In contrast, the silica content of the capsules investigated in Fig. 1 was below the detection limit.

Fig. 2 Scattering intensity plotted as ln(Iq²) vs.q² from incompletely dissolved silica-templated hollow capsules with 8 PE layers (●). The silica template diameter was about 1.2 µm. The solid and dashed lines denote the fits to the core–shell (n_c ≠ n_s) and core models, respectively.

3.2 Thickness and density of freestanding layers in capsules

Next, we measured the scattering from hollow capsules of 1.2 µm and 102 nm in diameter as a function of the layer number. Capsules fabricated from 8, 12 and 16 PE layers were compared. Fig. 3 refers to capsules templated on silica cores of 1.2 µm in diameter. The figure inset presents the ln(q²I(q)) vs.q representation employed for the layer thickness estimation by means of eqn (3). As expected, with increasing layer number the minimum shifts to smaller q values and the slope (see inset) becomes steeper. This is consistent with an increasing capsule wall thickness. The measured scattering intensity profiles were fitted and are displayed in Fig. 4 as the thickness per single layer, h_s, as a function of the total number of deposited polyelectrolyte layers. Fig. 4 further shows SFM thickness data inferred from dry collapsed hollow capsules with a diameter of 3 µm, as well as SANS thickness data corresponding to supported PEM on PS latex particles.⁵⁸

Fig. 3 Small-angle neutron scattering intensity profiles from 1.2 µm in diameter silica-templated hollow capsules with 8 (●), 12 (■) and 16 (▲) PE layers (from top to bottom). The solid lines denote the best fits. The inset shows the scattering intensity from the same experimental data plotted as ln(Iq²) vs.q². The solid lines in the inset denote linear fits at a small q.

Fig. 4 Average thickness per single layer of hollow capsules as a function of the total PE layer number. The capsule wall thickness data obtained from templating 102 nm (○) silica particles, 1.2 µm (●) silica particles and that of supported PEM on 168 nm PS latex particles⁵⁸ (■) were derived from SANS data. ▲ indicates thickness per single layer of 3 µm silica templated collapsed capsules derived from SFM data. The solid lines describe the linear fits to the thickness data. The horizontal line, located left at the h_s axis, denotes the corresponding thickness obtained from neutron reflectivity measurements of PEM deposited onto silica wafers in an aqueous environment.⁵⁸

As can be seen the thickness per single PE layer of the capsule wall does not depend on the diameter of the silica core. This is consistent with the absence of a curvature effect of supported layers, as demonstrated earlier.⁵⁸ Furthermore, the average single layer thickness decreases with the number of adsorbed PE layers both for supported and freestanding layers with approximately the same slope. This apparent decrease of the ratio of layer thickness to layer number can be related to a decreasing SLD value of the step function density profile of the shells, and, respectively, implies that the water content decreases with growing layer number. This assumption is based on the experimental fact, that the deposited mass itself, after the second pair of polyelectrolytes, in most cases increases linear with layer number.^3,4,49,78 An alternative explanation would be a slightly decreasing mass increment with layer number. Fig. 4 further reveals that the h_s value of dry hollow capsules as determined by SFM is only slightly smaller than the thickness of PEM supported on PS latex particles. This is consistent with dense supported layers with a water content of 42%.⁵⁸ The layers do not shrink notably upon drying. This comparison proves that, at least for empty capsules, SFM measurements on dried capsules provide a good estimate for the PEM layer thickness in solution.

Two observations made with capsules have attracted special interest. 1) The thickness of the freestanding layers in capsules was about 25% larger for all studied samples as compared with supported layers (Fig. 4). 2) The diameter of the capsule lumen was an almost exact match of the diameter of the bare silica particles, which were used as templates. Indeed, SANS data fitting yielded an average internal diameter of hollow capsules of about 104 nm, while the bar silica particles had a size of 102 nm as determined from the scattering data provided in Fig. 1. The water content of the freestanding layers was 52%, while in supported layers it was only 42%. Therefore, silica core dissolution is accompanied by a layer thickness increase and a corresponding density decrease.

3.3 Annealing response at high ionic strength

In view of the importance of annealing for permeability control we were interested to follow wall changes upon annealing. Small hollow capsules consisting of 8 PE layers were annealed at 70 °C for 4 h in salt solutions of two different ionic strengths, 0.1 and 1 mol L⁻¹NaCl. The corresponding scattering intensity profiles did not reveal any differences before and after annealing of the samples in the 0.1 mol L⁻¹NaCl solution. Layer thickness and size of the hollow capsules did not change. Remarkable changes in the SANS intensity were, however, observed if the annealing was conducted in 1 mol L⁻¹NaCl (Fig. 5). At q < 0.2 nm⁻¹, the positions of the scattering minima became shifted. This corresponded to a diameter of only 84 nm after temperature treatment. Annealing in 1 mol L⁻¹ thus induced capsule shrinking by about 20%.

Fig. 5 SANS intensity profiles from 102 nm silica-templated hollow capsules with 8 PE layers before (●) and after (■) temperature treatment at 70 °C during 4 hours in 1 mol L⁻¹NaCl solution. The solid lines denote the fits to the core–shell model. The inset shows the scattering intensity from the same experimental data plotted as ln(Iq²) vs.q². The solid lines denote linear fits at small q.

Furthermore, the position of the first minimum at higher q values corresponding to the layer thickness itself shifted toward smaller q, and the slope in the ln(q²I(q)) vs.q presentation (see inset) was correspondingly increased. This is consistent with a capsule wall thickness increase from 168 to 232 Å. Hence, the capsule wall became thicker by about 38% as a result of the annealing at high ionic strength.

Shell thickness and capsule diameters before and after temperature treatment can be converted into wall volume changes. We found that the volume of the wall decreased by about 13% as a result of the annealing in 1 mol L⁻¹NaCl. The water content of the capsule wall decreased to 45.7% after temperature treatment in 1 mol L⁻¹NaCl, which is quite similar to the water content of supported PE layers.⁵⁷ The observed silica-core-templated capsule shrinking and the corresponding densification of the wall correlates thus with SFM and SANS annealing studies conducted on melamine-core-templated³¹ and cell-templated capsules.^11,70

4. Conclusions

SANS provides a unique means of directly measuring the thickness of PEM capsule walls in water with high precision. Freestanding PEM films were thicker and had a higher water content than supported PE layers. The average single layer thickness of hollow capsules was independent of the surface curvature and decreased slightly with increasing PE layer number. SANS, furthermore, allows to detect traces of the core material within the capsule lumen. This technique thus can serve as a means to assess the quality of hollow capsules.

A remarkable result was that the diameter of the capsule interior coincided with the template size. This behaviour is different from that of melamine-templated capsules where an osmotically induced swelling during core dissolution results in capsules somewhat larger than the template size.³⁸ We further found that the capsule wall thickness is significantly larger than that of the multilayer originally deposited onto the silica cores. This finding together with the constancy of the lumen size has an interesting implication. It demonstrates that the increase in wall thickness occurring upon dissolution is strictly anisotropic. The wall expands only in the radial direction but not laterally. As it can be easily shown, a lateral expansion of the capsule wall comparable in extent to the increase in radial direction would result in a larger capsule lumen, which was, however, not observed. Without knowing the details of the kinetics of core dissolution one may only speculate about the reason for this interesting behaviour. A possible explanation would be that during the silica core dissolution large osmotic pressure differences are not built up, and, that the attachment of the bottom layer to the silica core surface prevents lateral expansion. The absence of lateral layer expansion in the case of silica cores is consistent with a smaller number of defects being created during capsule preparation, as compared with the case of melamine-templated capsules, where large pores and defects have been frequently reported.

The measurements of the annealed capsules showed that the induced transition of the wall topology was possibly achieved by simultaneously decreasing the electrostatic interactions between the oppositely charged layers constituents brought about by high ionic strength together with increasing the mobility of the layer constituents induced by the elevated temperatures.

Acknowledgements

We gratefully thank BENSC for providing experimental facilities and financial support. We acknowledge also the support from the MBF in the framework of the Nanobiotechnology programme 0312011C.

References

1. G. Decher, Science, 1997, 277, 1232.
2. E. Donath, D. Walther, V. N. Shilov, E. Knippel, A. Budde, K. Lowack, C. Helm and H. Möhwald, Langmuir, 1997, 13, 5294.
3. G. B. Sukhorukov, E. Donath, H. Lichtenfeld, E. Knippel, A. Budde and H. Möhwald, Colloids Surf., A, 1998, 137, 253.
4. E. Donath, G. B Sukhorukov, F. Caruso, S. A. Davis and H. Möhwald, Angew. Chem., Int. Ed., 1998, 37, 2202.
5. G. B. Sukhorukov, E. Donath, S. A. Davis, H. Lichtenfeld, F. Garuso, V. I. Popov and H. Möhwald, Polym. Adv. Technol., 1998, 9, 759.
6. S. Moya, G. B. Sukhorukov, M. Auch, E. Donath and H. Möhwald, J. Colloid Interface Sci., 1999, 216, 297.
7. C. Gao, S. Leporatti, E. Donath and H. Möhwald, J. Phys. Chem. B, 2000, 104, 7144.
8. S. Moya, E. Donath, G. B. Sukhorukov, M. Auch, H. Bäumler, H. Lichtenfeld and H. Möhwald, Macromolecules, 2000, 33, 4538.
9. S. Leporatti, A. Voigt, R. Mitlöhner, G. B. Sukhorukov, E. Donath and H. Möhwald, Langmuir, 2000, 16, 4059.
10. C. Gao, E. Donath, S.V. Moya,Dudnik and H. Möhwald, Eur. Phys. J. E, 2001, 5, 21.
11. S. Leporatti, C. Gao, A. Voigt, E. Donath and H. Möhwald, Eur. Phys. J. E, 2001, 5, 3.
12. C. Gao, S. Leporatti, S. Moya, E. Donath and H. Möhwald, Langmuir, 2001, 17, 3491.
13. R. Georgieva, S. Moya, S. Leporatti, B. Neu, H. Bäumler, C. Reichle, E. Donath and H. Möhwald, Langmuir, 2000, 16, 7075.
14. S. Moya, L. Dähne, A. Voigt, S. Leporatti, E. Donath and H. Möhwald, Colloids Surf., A, 2001, 183–185, 27.
15. S. Moya, R. Georgieva, H. Bäumler, W. Richter and E. Donath, Med. Biol. Eng. Comput., 2003, 41, 504.
16. S. Leporatti, C. Gao, A. Voigt, E. Donath and H. Möhwald, Eur. Phys. J. E, 2001, 5, 13.
17. A. Fery, F. Dubreuil and H. Möhwald, New J. Phys., 2004, 6, 1.
18. O. I. Vinogradova, O. V. Lebedeva and B. S. Kim, Annu. Rev. Mater. Res., 2006, 36, 143.
19. G. Sukhorukov, A. Fery and H. Möhwald, Prog. Polym. Sci., 2005, 30, 885.
20. G. B. Sukhorukov, A. Fery, M. Brumen and H. Möhwald, Phys. Chem. Chem. Phys., 2004, 6, 4078.
21. Z. Feng, Z. Wang, C. Gao and J. Shen, Adv. Mater., 2007, 19, 3687.
22. W. Tong and C. Gao, Colloids Surf., A, 2007, 295, 233.
23. R. v. Klitzing, J. E. Wong, W. Jaeger and R. Steitz, Curr. Opin. Colloid Interface Sci., 2004, 9, 158.
24. M. Gopinadhan, H. Ahrens, J. Günther, R. Steitz and C. Helm, Macromolecules, 2005, 38, 5228.
25. R. v. Klitzing, Phys. Chem. Chem. Phys., 2006, 8, 5012.
26. G. Ibarz, L. Dähne, E. Donath and H. Möhwald, Chem. Mater., 2002, 14, 4059.
27. T. Krebs, H. L. Tan, G. Andersson, H. Morgner and P. Gregory Van Patten, Phys. Chem. Chem. Phys., 2006, 8, 5462.
28. K. Köhler, P. M. Biesheuvel, R. Weinkamer, H. Möhwald and G. B. Sukhorukov, Phys. Rev. Lett., 2006, 97 , 188301.
29. S. Qin, D. Wei and X. , Jin, Macromol. Rapid Commun., 2006, 27, 11.
30. R. Müller, K. Köhler, R. Weinkamer, G. Sukhorukov and A. Fery, Macromolecules, 2005, 38, 9766.
31. K. Köhler, D. G. Shchukin, H. Möhwald and G. B. Sukhorukov, J. Phys. Chem. B, 2005, 109, 18250.
32. R. Georgieva, R. Dimova, G. Sukhorukov, G. Ibarz and H. Möhwald, J. Mater. Chem., 2005, 15, 4301.
33. I. McCullosch and S. W. Shalaby, Tailored Polymeric Materials for Controlled Delivery Systems, American Chemical Society, Washington, DC, 1998.
34. X. Qiu, S. Leporatti, E. Donath and H. Möhwald, Langmuir, 2001, 17, 5375.
35. L. Z. Zheng, X. Yao and J. H. Li, Curr. Anal. Chem., 2006, 2, 279.
36. M. Nolte and A. Fery, IEE Proc. Nanobiotechnol., 2006, 153, 112.
37. A. S. Angelatos, K. Katagiri and F. Caruso, Soft Matter, 2006, 2, 18.
38. C. S. Peyratout and L. Dähne, Angew. Chem., Int. Ed., 2004, 43, 3762.
39. X. Yang, X. Han and Y. Zhu, Colloids Surf., A, 2005, 264, 49.
40. G. B. Sukhorukov, A. L. Rogach, M. Garstka, S. Springer, W. J. Parak, A. Muñoz-Javier, O. Kreft, A. G. Skirtach, A. S. Susha, Y. Ramaye, R. Palankar and M. Winterhalter, Small, 2007, 3, 944.
41. G. B. Sukhorukov and H. Möhwald, Trends Biotechnol., 2007, 25, 93.
42. G. B. Sukhorukov, A. L. Rogach, B. Zebli, T. Liedl, A. G. Skirtach, K. Köhler, A. A. Antipov, N. Gaponik, A. S. Susha, M. Winterhalter and W. J. Parak, Small, 2005, 1, 194.
43. C. Graf and A. van Blaaderent, Langmuir, 2002, 18, 524.
44. L. Hao, C. Zhu, C. Chen, P. Kang, Y, Hu, W. Fan and Z. Chen, Synth. Met., 2003, 139, 391.
45. M. Prevot, A. L. Cordeiro, G. B. Sukhorukov, Y. Lvov, R. S. Besser and H. Möhwald, Macromol. Mater. Eng., 2003, 288, 915.
46. H. Wu, Z. Liu, X. Wang, B. Zhao, J. Zhang and C. Li, J. Colloid Interface Sci., 2006, 302, 142.
47. P. Schuetz and F. Caruso, Adv. Funct. Mater., 2003, 13, 929.
48. A. S. Angelatos, A. P. R. Johnston, Y. Wang and F. Caruso, Langmuir, 2007, 23, 4554.
49. A. P. R. Johnston, A. N. Zelikin, L. Lee and F. Caruso, Anal. Chem., 2006, 78, 5913.
50. C. Déjugnat, K. Köhler, M. Dubois, G. B. Sukhorukov, H. Möhwald, T. Zemb and P. Guttmann, Adv. Mater., 2007, 19, 1331.
51. Z. Feng, Z. Wang, C. Gao and J. Shen, Mater. Lett., 2007, 61, 2560.
52. M. Lösche, J. Schmitt, G. Decher, W, G. Bouwman and K. Kjaer, Macromolecules, 1998, 31, 8893.
53. R. Steitz, V. Leiner, R. Siebrecht and R. v. Klitzing, Colloids Surf., A, 2000, 163, 63.
54. R. v. Klitzing, J. E. Wong, W. Jaeger and R. Steitz, Curr. Opin. Colloid Interface Sci., 2004, 9, 158.
55. M. Gopinadhan, H. Ahrens, J. Günther, R. Steitz and C. Helm, Macromolecules, 2005, 38, 5228.
56. R. v. Klitzing, Phys. Chem. Chem. Phys., 2006, 8, 5012.
57. G. Tzvetkov, B. Graf, P. Fernandez, A. Fery, F. Cavalieri, G. Paradossi and R. H. Fink, Soft Matter, 2008, 4, 510.
58. I. Estrela-Lopis, S. Leporatti, S. Moya, A. Brandt, E. Donath and H. Möhwald, Langmuir, 2002, 18, 7861.
59. D. Qiu, C. Flood and T. Cosgrove, Langmuir, 2008, 24, 2983.
60. J. C. Marshall, T. Cosgrove, F. Leermakers, T. M. Obey and C. A. Dreiss, Langmuir, 2004, 20, 4480.
61. P. J. Dale, B. Vincent, T. Cosgrove and J. Kijlstra, Langmuir, 2005, 21, 12244.
62. C. Flood, T. Cosgrove, D. Qiu, Y. Espidel, I. Howell and P. Revell, Langmuir, 2007, 23, 2408.
63. M. Zackrisson, A. Stradner, P. Schurtenberger and J. Bergenholtz, Langmuir, 2005, 21, 10835.
64. A. Rübe, G. Hause, K. Mäder and J. Kohlbrecher, J. Controlled Release, 2005, 107, 244.
65. J. Oberdisse, Curr. Opin. Colloid Interface Sci., 2007, 12, 3.
66. K. Matsumoto, H. Hasegawa, T. Hirabayashi, T. Harada and H. Matsuoka, Polym. Prepr., Jpn., 2005, 54, 2820.
67. J.-F. Berret and J. Oberdisse, Physica B (Amsterdam), 2004, 350, 204.
68. M. Sotiropoulou, F. Bossard, E. Balnois, J. Oberdisse and G. Staikos, Langmuir, 2007, 23, 11252.
69. E. Serefoglou, J. Oberdisse and G. Staikos, Biomacromolecules, 2007, 8, 1195.
70. G. Carrot, A. E. Harrak, J. Oberdisse, J., Jestin and F. Boué, Soft Matter, 2006, 2, 1043.
71. I. Estrela-Lopis, S. Leporatti, E. Typlt, D. Clemens and E. Donath, Langmuir, 2007, 23, 7209.
72. P. Strunz, J. Šaroun, U. Keiderling, A. Wiedenmann and R. Przenioslo, J. Appl. Crystallogr., 2000, 33, 829.
73. S. Kline, J. Appl. Crystallogr., 2006, 39, 895.
74. J. S. Pedersen, Adv. Colloid Interface Sci., 1997, 70, 17.
75. S. Förster and C. Burger, Macromolecules, 1998, 31, 879.
76. J. S. Pedersen, D. Posselt and K. Mortensen, J. Appl. Crystallogr., 1990, 23, 321.
77. J. G. Barker and J. S. Pedersen, J. Appl. Crystallogr., 1995, 28, 105.
78. V. Bosio, F. Dubreuil and A. Fery, Colloids Surf., A, 2004, 243, 147.

---