Silica nanoparticles with a hybrid organic–inorganic shell

Abstract

Non-agglomerated nanoparticles of an average diameter of 200 nm with a hybrid organic–inorganic shell were prepared by reacting the COOH groups of poly(N-dicarbazolyl-lysine)-covered silica nanoparticles with zirconium tert-butoxide, Zr(OtBu)₄, followed by sol–gel processing.

---

Introduction

Various strategies are being developed to prepare nanoparticles of increasingly complex structures and morphologies, such as multi-shell, Janus, or hollow particles. Illustrative examples are given in ref. 1–9. A prominent role is played by inorganic particles covered by organic species. Bi- or multi-layer hybrid organic–inorganic particles with polymeric shells can be prepared by different approaches, such as reverse micelle systems,¹⁰ heterophase polymerizations,⁴ or surface-initiated atom transfer radical polymerization processes.^11–13 In that case, organic shells are generally formed around the inorganic core of the nanoparticle. An inverse approach may also be used, i.e. growing inorganic nanoparticles inside dendrimeric nanoreactor voids for particle containment and passivation.^14–17

The great majority of known hybrid inorganic–organic nanoparticles contain only one kind of inorganic matter, e.g.titania, silica or gold, and there are only a few reports on hybrid nanoparticles possessing two different inorganic constituents. For example, titania nanoparticles were generated by sol–gel processing in hyperbranched polymers substituted by pendant (CH₂)₃Si(OEt)₃groups. The Si(OEt)₃ groups served as mediator functionalities between the titania nanoparticles and the polymeric shell and, consequently, do not constitute an intermediate inorganic shell.¹⁸ SiO₂/polystyrene/TiO₂ multilayer core–shell hybrid microspheres were obtained by mixing positively charged SiO₂/polystyrene hybrid particles with Ti(OtBu)₄ followed by sol–gel reaction.¹⁹Polystyrene is an inert polymer that does not interact with the metal alkoxide precursor. Thus, polymers equipped with pendant functional groups capable of coordinating metal alkoxides should be beneficial for the formation of an anchored secondary inorganic structure. A similar approach was previously used for the generation of a gold metal layer on bilayer gold/silica nanoparticles. The outer gold layer was obtained by interaction of a gold hydroxide aqueous solution with amine groups of (3-aminopropyl)trimethoxysilane grafted onto the silica layer followed by reduction.²⁰ A three-layer Au/silica/polyisoprene nanocomposite was prepared by atom transfer radical polymerization from the surface of Au@SiO₂nanoparticles.¹³ A different sequence of the layers was obtained by reacting dodecanethiol-stabilized gold nanoparticles with octadecyldimethyl(3-trimethoxysilylpropyl)ammonium chloride. Hydrophobic interactions between the two long alkyl chains of both the adsorbed thiol and the ammonium compound resulted in a self-assembled structure where the alkoxysilyl groups were located on the outer surface of the hybrid nanoparticles. The alkoxysilyl groups were converted in a silica shell around the organic-covered gold nanoparticles by hydrolytic polycondensation.²¹

In the present work, we describe the preparation of a morphologically different three-component nanoparticle system composed of a zirconia adlayer on (S)-N-dicarbazolyl-lysine (polyDCL)-coated silica nanoparticles. The functional polyDCL contains the necessary COOH groups for coordinating Zr(OR)_x groups. Following sol–gel processing, the resulting spherical hybrid inorganic–organic–inorganic nanosized particles of a 200 nm average diameter are built on two inorganic components, an inorganic SiO₂ core and an outer mixed hybrid zirconia/polyDCL adlayer.

Experimental

1. Measurements

Dynamic light scattering (DLS) measurements. For the measurement the solid was dissolved in ethanol. The DLS experiments were carried out without previous sonication of the samples. The run time of the measurements was 10 s. Every size distribution curve was obtained by averaging 10 measurements.

The apparatus was an ALV/CGS-3 compact goniometer system, equipped with ALV/LSE-5003 light scattering electronics and multiple τ digital correlator, and a 632.8 nm JDSU laser 1145P.

Transmission electron microscopy (TEM) measurements. Samples for transmission electron microscopy measurements were prepared by dispersing the particles in ethanol prior to deposition on a carbon coated TEM Cu grid. TEM measurements were performed on a 100 kV JEOL JEM-100CX (USTEM, Vienna University of Technology).

TEM associated with energy dispersive X-ray spectroscopy (EDX) measurements. Sample measurements were prepared by dispersing the particles in ethanol prior to deposition on a carbon coated TEM Cu grid. TEM measurements were performed on a 200 kV FEI TECNAI F20 S-TWIN apparatus with field emission source connected to an EDX detector (USTEM, Vienna University of Technology).

Fourier transform infra-red (FT-IR) spectrum. The products were pelletized in KBr before measurement. The spectrometer was a Bruker Tensor-27-DTGS equipped with an Interferometer RockSolid™ and a DigiTect™ detector system, high sensitivity DLATGS, piloted by the software OPUS™.

SEM . The scanning electron microscope used in this work was a Philips 200 kV ESEM-FEG XL30.

2. Synthesis

Typical protocol for the preparation of nanosized SiO₂/polyDCL composite particles. The carbazole based monomer (S)-2,4-di(9H-carbazol-9-yl)butanoic acid (DCL, 223.5 mg, 0.5 mmol) was dissolved in CH₃COCH₃ (5.0 mL) using smooth sonication (Bransonic cleaning bath). The DCL acetone solution was slowly added to an aqueous dispersion of commercially available 50 nm-sized SiO₂nanoparticles (Nyacol^® DP5480, 30 wt% in ethylene glycol, 200.0 mg, 4.8 mL of H₂O), followed by the (NH₄)₂S₂O₈ oxidant (APS, 570.0 mg, 2.5 mmol). The reaction was ultrasonicated for 1 h using a VCX-750 sonicator operating at 40% power output. At reaction completion, the medium was centrifuged (10 000 rpm, 10 min at 4 °C). The solid precipitate was washed five times using a sequential redispersion–centrifugation cycle using a 1 : 1 v/v CH₃COCH₃–H₂O washing mixture (10 mL). Resulting SiO₂/polyDCL-based composite nanoparticles were dried in high vacuum (10⁻⁴ mm Hg, 8 h) before full characterization using a combination of analytical (elemental analysis, TGA/DSC, spectroscopic (FT-IR)) and microscopic (SEM and TEM) methods.

Hybrid SiO₂/polyDCL-based nanoparticles contained 87.5 wt% of a polymeric polycarboxylated polyDCL shell for an average 100 nm diameter.

Typical protocol for the preparation of SiO₂/polyDCL/ZrO₂ composite nanoparticles. The synthesis was carried out under argon flux. 44 mg of polyDCL-capped silica nanoparticles were dispersed in 40 mL freshly distilled ethanol. 75 mg of an 80 wt% solution of Zr(OtBu)₄ in nBuOH (i.e. 60 mg Zr(OtBu)₄, 0.157 mmol) were added at room temperature, under stirring, to the nanoparticle dispersion. Subsequently four equivalents of distilled water (0.628 mmol, 18 mg) and 0.01 equivalent of acetic acid per mol of Zr(OtBu)₄ (100 μL) were added to the solution. The stirring was maintained overnight at room temperature.

Protocol for dissolving the core of SiO₂/polyDCL/ZrO₂ composite nanoparticles. The particles were treated during 3 h under stirring with a 32 wt% aqueous solution of hydrofluoric acid at room temperature.

Results and discussion

The starting silica/polyDCL nanosized composites are new composites based on both silica and the conducting polymer polyDCL (Scheme 1). They were prepared using an ultrasound-assisted chemical oxidation of the monocarboxylated (S)-N-dicarbazolyl-lysine monomer (DCL) in the presence of 50 nm-sized SiO₂nanoparticles (1 : 1 v/v acetone–H₂O, (NH₄)₂S₂O₈ oxidant, 1 h, 55–60 °C).²² The resulting hybrid SiO₂/polyDCL-based nanoparticles were composed of 87.5 wt% of an insoluble polycarboxylated polyDCL shell with an average 80–150 nm particle diameter, centered on 100 nm (Fig. 1). All particle size and size distribution curves were obtained by averaging ten dynamic light scattering (DLS) measurements.

Fig. 1 DLS graph of the size and size distribution of the starting SiO₂/polyDCL nanoparticles (dispersion in ethanol).

Scheme 1 Synthesis strategy for the preparation of silica/polyDCL/zirconia nanoparticles.

The FT-IR spectrum of the starting SiO₂/polyDCL nanoparticles (Fig. 2, top) shows peaks relating to the different bonds and functional groups that characterize polyDCL. The band around 3490 cm⁻¹ corresponds to polymer O–H groups, while both 3050 and 2940 cm⁻¹ vibration bands relate to N–H and C–H groups respectively. C=O vibrations are located at 1720 cm⁻¹ (indicated with an arrow on Fig. 2, top), while the peak at 1600 cm⁻¹ indicates the presence of C=C double bonds. Vibrations corresponding to the silica matrix can be found at 1460 and 1330 cm⁻¹. The presence of the strong band at 1720 cm⁻¹ proves the presence of the carboxylic acid functionality within the polyDCL shell, which is strongly adsorbed onto the SiO₂ particle surface.

Fig. 2 FT-IR spectra of SiO₂/polyDCL (top), and SiO₂/polyDCL/ZrO₂ (bottom) nanoparticles.

Reaction of metal alkoxides with carboxylic acids is a well-known method to obtain organically modified precursors for sol–gel processing.²³ Although the initial reaction of Zr(OR)₄ with carboxylic acids often results in the formation of carboxylate-capped oxo/hydroxo clusters,²⁴ sol–gel processing provides carboxylate-substituted zirconia entities. In the work presented in this article, the COOH groups of the polyDCL shell of the starting composite nanoparticles were exploited to further coordinate the zirconium center (Scheme 2) prior to sol–gel processing. The synthesis procedure is shown in Scheme 1.

Scheme 2 Coordination of a carboxylic acid group onto a metal center M.

PolyDCL-capped silica nanoparticles, dispersed in freshly distilled ethanol (1.1 mg of nanoparticles per mL of ethanol for a total volume of 40 mL), were added to an 80 wt% solution of Zr(OtBu)₄ in nBuOH (1.7 g solution of zirconium alkoxide per g of dispersed nanoparticles, i.e. 1.4 g of zirconium alkoxide per g of dispersed nanoparticles) in an inert argon atmosphere. Four equivalents of water and 0.01 equivalent of acetic acid per mol of Zr(OtBu)₄ were added to this mixture under stirring.

Reaction of 100 nm average diameter-sized polyDCL-capped silica nanoparticles with the zirconium alkoxide, followed by hydrolysis, led to spherical nanoparticles of increased diameters. As a matter of fact, particles with diameters ranging from 100 to 400 nm, centered at 200 nm, were observed by DLS (dispersion in ethanol, Fig. 3). The observed increase of the particle diameter by a factor of 1.5 to 2 indicates that the reaction of polyDCL-capped silica particles with the zirconium alkoxide led to a new composite material consisting of ZrO₂, formed by sol–gel processing and attached to the polyDCL-capped silica particles.

Fig. 3 DLS graph of the size and size distribution of the new silica/polyDCL/zirconia particles (dispersion in ethanol).

The time-dependent colloidal stability was evaluated by DLS after three months. The modified nanoparticles were stored at room temperature in the 40 mL ethanol solution used for the synthesis. Size and size distribution features were unchanged (Fig. 4).

Fig. 4 DLS graph of the size and size distribution of the silica/polyDCL/zirconia nanoparticles after 3 months.

The efficiency of the metal alkoxide coordination by the polyDCL carboxylic acid groups (Scheme 2) can be highlighted by comparison of the FT-IR spectra of samples before and after modification by sol–gel processing. The modified three-component nanoparticles, SiO₂/polyDCL/ZrO₂, show merged bands corresponding to both silica and zirconia at 1460 and 1330 cm⁻¹. In addition, newly formed Zr–OH bonds arising from sol–gel processing cause a broadening of the 3490 cm⁻¹ band. Except the C=O band at 1720 cm⁻¹, all the other bands characteristics of the polyDCL are still observable (O–H groups around 3490 cm⁻¹, 3050 and 2940 cm⁻¹ vibration bands relating to N–H and C–H groups respectively, the peak at 1600 cm⁻¹ indicating the presence of C=C double bonds). The disappearance of the strong peak at 1720 cm⁻¹ (located by an arrow in Fig. 2, top) indicates the efficient coordination of polyDCL carboxylic acid groups onto the Zr metal center (Fig. 2, bottom).

The size, and size distribution of the resulting three-compound nanoparticles indicated by DLS measurements was confirmed by TEM (Fig. 5). No condensation between particles was observed, i.e. hydrolytic polycondensation of polycarboxylate-substituted Zr(OtBu)₄ occurs only within the polymeric shell of the silica/polyDCL particles. The presence of both SiO₂ and ZrO₂ components in the new hybrid particles was also proven by EDX analysis (Fig. 5 left, at three positions denoted 1–3). The illustrative EDX spectrum of position 3 is disclosed in Fig. 5 (right). From these observations, it clearly appears that each nanoparticle is composed of both silica and zirconia structures.

Fig. 5 (left) TEM photomicrograph of the silica/polyDCL/zirconia nanoparticles; (right) EDX spectrum of position 3 in the same TEM photomicrograph.

Moreover, an EDX line scan profile of both Si and Zr elemental distributions across one of the particles (Fig. 6) showed that Si and Zr are present at a relatively constant level. Accordingly, the formation of an outer zirconia shell around the polyDCL-covered silica particles (Scheme 1, top) may be doubtful. Zirconia particulates embedded in the polymeric shell, i.e. resulting in a mixed organic–inorganic outer shell, have to be also considered (Scheme 1, bottom). Nevertheless, the EDX measurement is made on the indicated line but analyzes the entire thickness of the deposited product, thus the constant level of silicon and zirconium elements detected could be a consequence of particles overlapping. The obtained zirconia structures are amorphous, as usually observed for sol–gel materials, and thus show no diffraction pattern.

Fig. 6 (left) TEM photomicrograph of the silica/polyDCL/zirconia nanoparticles; (right) elemental EDX line scan analysis profile of line 1.

When an ethanol dispersion of the obtained silica/polyDCL/zirconia nanoparticles was coated on a glass plate and dried at room temperature under air, a multi-layered film of agglomerated nanoparticles was obtained. SEM analysis (Fig. 7) shows a narrow average size and size distribution of the nanoparticles, as also observed by DLS. Nanoparticle diameters from the TEM range from 130 to 300 nm, centered at 200 nm.

Fig. 7 SEM of three-component silica/polyDCL/zirconia nanoparticles.

The silica/polyDCL/zirconia nanoparticles were treated with a 32 wt% aqueous solution of hydrofluoric acid in order to dissolve the silica core. After 3 h treatment at room temperature, the silica core was completely dissolved and small zirconia particles were observed (Fig. 8).

Fig. 8 TEM photomicrograph before (left) and after (right) 3 h of HF treatment of silica/polyDCL/zirconia nanoparticles.

They appeared almost cubic although the zirconia phase was amorphous (XRD). The size distribution of the observed 150 nm-sized ZrO₂nanoparticles is rather broad (120–250 nm). Their appearance did not allow any definite conclusion to be drawn about whether they constituted part of a broken zirconia shell or isolated nanoparticles originally embedded in the polymeric polyDCL shell.

In conclusion, hybrid organic–inorganic SiO₂/polyDCL/ZrO₂nanoparticles possessing a complex dual inorganic composition were prepared. They were composed of a silica core and an organic polyDCL layer with pendant carboxylic acid groups. The latter enabled further reaction with a metal alkoxide precursor that resulted in zirconia nanoparticulates after sol–gel processing. The feasibility of this procedure was demonstrated for one zirconium alkoxide precursor, Zr(OtBu)₄, but might be readily applicable for equivalent nanosized systems based on different inorganic cores and metal alkoxides.

There are two structural possibilities of how zirconia may attach to polyDCL-coated silica nanoparticles. The first is a zirconia shell covering the silica/polyDCL-capped particles (Scheme 1, top). The second possibility will lead to the formation of a mixed organic (polyDCL)–inorganic (ZrO₂) shell (Scheme 1, bottom) deposited around the silica core. The first possibility would require that a significant part of the COOH groups of the polyDCL polymer would be located on the outer surface of the starting hybrid silica/polyDCL nanoparticles. Given the employed ratios used between silica nanoparticles and the zirconium alkoxide precursor, the zirconia shell built around the silica/polyDCL-capped nanoparticles should be very thin. However, the increase of the particle diameter after sol–gel processing is more indicative of the alternative hybrid structure, which is also supported by elemental EDX analysis and dissolution studies. The formation of such a hybrid polyDCL/zirconia hybrid shell around the silica core is also more likely from a chemical point of view, since the COOH groups of the polyDCL shell are not only present onto the particle surface, but also within the organic shell.

Acknowledgements

This work was supported by the European Commission in the frame of the 6th Framework Integrated Project NAPOLYDE (Contract No. NMP2-CT-2005-515846). The authors thank Emmanuel Scolan, CSEM Switzerland, for SEM investigations.

References

1. M. Green, Small, 2005, 1, 684.
2. D. Lee, D. Omolade, R. E. Cohen and M. F. Rubner, Chem. Mater., 2007, 19, 1427.
3. A. Perro, S. Reculusa, S. Ravaine, E. Bourgeat-Lami and E. Duguet, J. Mater. Chem., 2005, 15, 3745.
4. S. Reculusa, C. Poncet-Legrand, S. Ravaine, C. Mingotaud, E. Duguet and E. Bourgeat-Lami, Chem. Mater., 2002, 14, 2354.
5. S. Sadasivan and G. B. Sukhorukov, J. Colloid Interface Sci., 2006, 304, 437.
6. M. Sastry, A. Swami, S. Mandal and P. R. Selvakannan, J. Mater. Chem., 2005, 15, 3161.
7. A. van Blaaderen, Surfactant Sci. Ser., 2006, 131, 233.
8. Y.-Y. Yang, Y. Wang, R. Powell and P. Chan, Clin. Exp. Pharmacol. Physiol., 2006, 33, 557.
9. Functional Coatings by Polymer Microencapsulation, ed. S. K. Ghosh, Wiley-VCH Verlag GmbH & Co., Weinheim, 2006.
10. D. Shukla and A. Mehra, Langmuir, 2006, 22, 9500.
11. Z. Lei and S. Bi, Mater. Lett., 2007, 61, 3531.
12. D. Holzinger and G. Kickelbick, Chem. Mater., 2003, 15, 4944.
13. D. Holzinger, L. M. Liz-Marzan and G. Kickelbick, J. Nanosci. Nanotechnol., 2006, 6, 445.
14. F. N. Crespilho, M. E. Ghica, V. Zucolotto, F. C. Nart, O. N. Oliveira, Jr. and C. M. A. Brett, Electroanalysis, 2007, 19, 805.
15. F. N. Crespilho, F. C. Nart, O. N. Oliveira and C. M. A. Brett, Electrochim. Acta, 2007, 52, 4649.
16. Y. Haba, C. Kojima, A. Harada, T. Ura, H. Horinaka and K. Kono, Langmuir, 2007, 23, 5243.
17. Y. Zhu, H. Zhu, X. Yang, L. Xu and C. Li, Electroanalysis, 2007, 19, 698.
18. A. Di Gianni, S. Trabelsi, G. Rizza, M. Sangermano, H. Althues, S. Kaskel and B. Voit, Macromol. Chem. Phys., 2007, 208, 76.
19. J. Zhou, M. Chen, X. Qiao and L. Wu, Langmuir, 2006, 22, 10175.
20. X. Xia, Y. Liu, V. Backman and G. A. Ameer, Nanotechnology, 2006, 17, 5435.
21. H. Fan, Z. Chen, C. J. Brinker, J. Clawson and T. Alam, J. Am. Chem. Soc., 2005, 127, 13746.
22. E. Kuperschmidt, Preparation and Characterization of Nanocomposites Based on SiO₂ and SiC Nanoparticles and Functional Polypyrrole/polycarbazole Polymers, MSc Thesis, Bar-Ilan University, Ramat-Gan, Israel, 2007.
23. U. Schubert, J. Sol–Gel Sci. Technol., 2003, 26, 47.
24. U. Schubert, Chem. Mater., 2001, 13, 3487.

---