666
Langmuir 2008, 24, 666-669
Directed Synthesis of Stable Large Polyoxomolybdate Spheres Soumyajit Roy,* Lydia C. A. M. Bossers, Hans J. D. Meeldijk, Bonny W. M. Kuipers, and Willem K. Kegel* Van’t Hoff Laboratory for Physical and Colloid Chemistry, Debye Institute, and Electron Microscopy Department of Molecular Cell Biology, UniVersity of Utrecht, Padualaan 8, 3584 CH, Utrecht, The Netherlands ReceiVed NoVember 6, 2007. In Final Form: December 17, 2007 Polyoxometalates or POMs, a class of inorganic transition metal-oxide based clusters, have gained significant interest owing to their catalytic, magnetic, and material science applications. All such applications require high surface area POM based materials. However, chemically synthesized POMs are still at most in the range of a few nanometers, with their size and morphology being difficult to control. Hence, there is an immediate need to develop design principles that allow easy control of POM morphology and size on mesoscopic (50-500 nm) length scales. Here, we report a design strategy to meet this need. Our method reported here avoids a complex chemical labyrinth by using a prefabricated cationic 1,2-dioleol-3-trimethylammonium-propane (DOTAP) vesicle as a scaffold/structure directing agent and gluing simple anionic heptamolybdates by electrostatic interaction and hydrogen bonds to form large POM spheres. By this method, complexity in the resulting structure can be deliberately induced either via the scaffold or via the oxometalate. The high degree of control in the matter of the size and morphology of the resulting POM superstructures renders this method attractive from a synthetic standpoint.
Introduction Polyoxometalates or POMs, a class of inorganic transition metal-oxide based clusters, have gained significant interest owing to their catalytic, magnetic, and material science applications.1,2 All such applications require high surface area POM based materials.3,4 However, chemically synthesized POMs are still at most in the range of a few nanometers, with their size and morphology being difficult to control. Hence, there is an immediate need to develop design principles that allow easy control of POM morphology and size on mesoscopic (50-500 nm) length scales. Here, we report a design strategy to meet this need (Figure 1). At this point, it is worth mentioning that a significant development in the field of POMs was achieved when complex molecular POMs were self-assembled as large hollow spheres of 50-100 nm by the Liu group.5-8 Our method reported here augments such ongoing efforts for large POM design using a more controlled approach. It uses a prefabricated cationic vesicle as a scaffold/structure directing agent and glues simple anionic oxomolybdates by electrostatic interaction and hydrogen bonds to form large POM spheres. By this method, complexity in the resulting structure can be deliberately induced either via the scaffold or via the oxometalate. The high degree of control in the matter of the size and morphology of the resulting POM superstructures renders this method attractive from a synthetic standpoint. For instance, the option of varying the overall POM topology just by changing the shape of the vesicle exists and can be explored. * To whom correspondence should be addressed. E-mail:
[email protected] (S.R.);
[email protected] (W.K.K.). S.R. is currently at BASFISIS, Strasbourg, France. Fax: (+31) 30-253-3870. (1) Pope, M. T. Heteropoly and isopoly oxometalates; Springer: Berlin, 1983. (2) Pope, M. T.; Mu¨ller, A. Polyoxometalate chemistry: From topology Via self-assembly to applications; Kluwer Academic Publishers: Dordrecht, London, 2001. (3) Hill, C. L. Chem. ReV. 1998, 98 (1). (4) Pope, M. T.; Yamase, T. Polyoxometalate chemistry for nano-composite design; Kluwer Academic/Plenum Publishers: New York, London, 2002; pp 1-235. (5) Liu, T. J. Am. Chem. Soc. 2002, 124 (37), 10942-10943. (6) Liu, T. J. Am. Chem. Soc. 2004, 126 (1), 406. (7) Liu, T. J. Am. Chem. Soc. 2003, 125 (2), 312-313. (8) Liu, G.; Liu, T. Langmuir 2005, 21 (7), 2713-2720.
To date, cationic lipid vesicles have been used in combination with polymers and polyelectrolytes to exploit their very rich phase behavior for developing new materials.9,10 They have further been used as drug delivery agents,11,12 in combination with DNA,13 for delivering boron to cure rheumatoid arthritis,14 in cosmetics manufacturing,9 and so forth. They have also been used as templates for making silicone nanocapsules,15 porous lamellar silica,16 and other core-shell particles. For an overview on such core-shell particles, see ref 17. Here, we explore and extend such possibility of scaffold based design for forming large POM spheres in a controlled fashion. The synthesized spheres have been characterized by dynamic light scattering (DLS), transmission electron microscopy (TEM), cryo-TEM, energy dispersive X-ray analyses (EDX) coupled with scanning TEM with a high angle annular dark field detector (STEM/HAADF), Fourier transform infrared (FTIR), Raman spectroscopy, and electrophoretic mobility measurements. Materials and Methods Synthetic Procedure. Ammonium heptamolybdate tetrahydrate ((NH4)6[Mo7O24]‚4H2O, MW ) 1235.86 g/mol) and DOTAP (1,2dioleol-3-trimethylammonium-propane, MW ) 710.34 g/mol) were obtained as powders from J. T. Baker and Avanti Lipids, respectively, and were used without further purification. DOTAP vesicles were prepared following standard procedure by drying overnight 100 µL of 17 mg/mL DOTAP in a 1:1 chloroform/methanol mixture followed by hydration in 10 mL of water and 90 min sonication in an ultrasonic (9) Lasic, D. D. Trends Biotechnol. 1998, 16 (7), 307-321. (10) Lasic, D. D. Liposomes: from physics to applications; Elsevier: Amsterdam, New York, 1993. (11) Lasic, D. D.; Papahadjopoulos, D. Medical applications of liposomes; Elsevier: Amsterdam and Oxford, 1998. (12) Swaan, P. W.; Szoka, F. C., Jr.; Øie, S. AdV. Drug DeliVery ReV. 1996, 20 (1), 59-82. (13) Lasic, D. D.; Barenholz, Y. Handbook of nonmedical applications of liposomes. From gene deliVery and diagnostics to ecology. CRC: Boca Raton, FL, 1996; Vol. 4. (14) Watson-Clark, R. A.; Banquerigo, M. L.; Shelly, K.; Hawthorne, M. F.; Brahn, E. Proc. Natl. Acad. Sci. U.S.A. 1998, 95 (5), 2531-2534. (15) Kepczynski, M.; Ganachaud, F.; He`mery, P. AdV. Mater. 2004, 16 (20), 1861-1863. (16) Tanev, P. T.; Pinnavaia, T. J. Science 1996, 271 (5253), 1267-1269. (17) Caruso, F. AdV. Mater. 2001, 13 (1), 11-22.
10.1021/la703467d CCC: $40.75 © 2008 American Chemical Society Published on Web 01/09/2008
Letters
Langmuir, Vol. 24, No. 3, 2008 667
Figure 1. Schematic representation of the design strategy for making of POM large spheres. The spherical DOTAP template (left) and DOTAP-scaffold POM supersphere (right) are shown schematically with surface corrugation. A plausible H-bonding between the DOTAP and the binding heptamolybdate (in wire-frame representation with ModO groups pointing toward the CdO) is implied. (See text.) bath. To this dispersion, a concentration range of ammonium heptamolybdate was added (0.1-10 µmol). The pH of these dispersions was ∼5.5. Of this series, the concentration of ammonium heptamolybdate in the range of 1-2 µmol was found to induce instability in the dispersions. The pH of the dispersions for the purpose was measured with a Schott-Gera¨te pH meter. Characterization of the Material. Dynamic Light Scattering (DLS) Experiments and Analyses. DLS measurements were performed using an argon laser (Spectra Physics, Series 2020; λ ) 514.5 nm) operating at a laser intensity of 450 mW and at a temperature of 298 K. The samples were measured in Danliker cuvettes. The intensity autocorrelation functions (IACF) were obtained using a Malvern correlator. Each DLS measurement was performed at six angles (25°-135°). At each angle, 10-20 decay curves were measured and the decay exponent was obtained by fitting the decay curves to a first-order cumulant fit with only DOTAP vesicles and various M/D molar concentrations. The radii range of the particles obtained from the range of corresponding apparent translational diffusion coefficients with varying M/D ratios are discussed later. Cryo-Transmission Electron Microscopy and EDX/STEM. The characterization of the composite was primarily done by the use of a Tecnai 20 transmission electron microscope (FEI Company) operated at an accelerating voltage of 200 kV. The imaged particles revealed characteristic features discussed in a later section. Further confirmation was carried out by EDX/STEM-HAADF (energy dispersive X-ray/scanning transmission electron microscopy with a high angle annular dark field detector) using a Tecnai 20 EDX/TEM detector. The cryo-TEM images were also obtained using the same microscope with the use of a Gatan cryo-holder. The TEM micrographs have been processed using SIS software (Soft Imaging System), and the EDX spectra were acquired using TIA software (Tecnai imaging and analysis software). FTIR Spectroscopy and Raman Spectroscopy. FTIR measurements were carried out at room temperature on a Perkin-Elmer 2000 Fourier transform spectrometer after drying the dispersions and then with KBr as pellets. Spectra were recorded with an attenuated total reflection (ATR) accessory (PIKE) equipped with a diamond crystal as the reflecting element. Data point resolution of the spectra was 4 cm-1, and 10 scans were accumulated for one spectrum. Raman spectroscopic measurements were carried out on a Kaiser RXN spectrometer equipped with a 50 mW 532 nm diode laser for excitation, a holographic grating for dispersion, and a Peltier-cooled Andor charge-coupled device (CCD) camera for detection. Spectra were recorded in glass vials at room temperature. The detector pixel resolution was about 2 cm-1, and 10 scans were accumulated for one spectrum at an exposure time of 20 s per scan.
Electrophoretic Mobility. The electrophorectic mobilities were measured with a Coulter DELSA 440 SX instrument using dilute samples of the DOTAP dispersion and the composite in water at a pH of ∼5.5 and a temperature of 298 K. The electrophorectic mobilities of the composite showed charge inversion with respect to the starting DOTAP and are discussed in the next section.
Results and Discussion The synthesis was carried out by adding an appropriate amount of heptamolybdate to a previously prepared DOTAP vesicle (following standard practice; for details, see the experimental section). (Also note that for convenience heptamolybdate has been denoted as molybdate in this article from here onward.) Interestingly, there is a window of molybdate/DOTAP (M/D) concentration for which a stable dispersion is formed. However, beyond this window, the dispersion becomes unstable and then stable again. The phenomenon of the formation of a stableunstable-stable dispersion was followed by electrophoretic mobility measurements, and it points to the operation of a charge inversion mechanism18 as a function of varying M/D concentration which may be explained as follows (Figure 2). DOTAP vesicles are positively charged. Upon addition of anionic molybdates to this dispersion, the positive charge on the vesicles is reduced and finally instability is induced in a certain concentration window (1.5 > M/D > 0.6). The dispersion becomes nearly neutral and aggregates. If all the added POMs reside at the vesicles, this instability should manifest at M/D ) 0.16, but the much higher observed values indicate the presence of free molybdates in the dispersion. Consequently, an additional amount of molybdate is necessary for reaching the unstable regime. However, upon further addition of molybdates (10 > M/D > 3), the dispersion undergoes charge inversion and is now stabilized by negative charges. From this ratio, the interface structure of the synthesized POM giant spheres can immediately be deduced and is discussed later. We now investigate the nature of this stable composite. DLS investigation of the resulting dispersion (assuming spherical topology) implies the presence of particles with hydrodynamic radii in the range of 80-90 nm comparable to the radii of the starting DOTAP vesicles, calculated from the (18) Grosberg, A. Y.; Nguyen, T. T.; Shklovskii, B. I. ReV. Mod. Phys. 2002, 74 (2), 329-345.
668 Langmuir, Vol. 24, No. 3, 2008
Letters
Figure 2. Charge inversion in the course of POM-DOTAP sphere formation. Plot of electrophoretic mobility (U) with molybdate/ DOTAP (M/D) concentration reveals an inversion of charge from +5 µm cm V-1 s-1 for very low M/D ratio (very low molybdate concentration) to -5 µm cm V-1 s-1 for very high M/D ratio (very high molybdate concentration, M/D ≈ 3 and higher). The regions of instability are highlighted. (M/D represents molar ratio.)
corresponding diffusion coefficients obtained from the intensity autocorrelation functions (IACF). The radii of the particles have been calculated using the Einstein-Smoluchowski equation assuming spherical topology. The results of DLS investigations with varying M/D concentration are tabulated below. Further investigations with electron microscopy followed.
Cryo-TEM on the dispersions of the composite showed spherical particles with dark ridges implying the presence of (heavier) molybdates on the surfaces of the DOTAP vesicles. Additionally, aggregates of those spheres with radii of 60-100 nm were found to coexist with single discrete spheres of smaller radii (∼20-50 nm) (Figure 3a). Upon drying the dispersion of the composite, two interesting effects were observed in TEM. First, no significant shrinkage in the diameter of the composite was observed upon drying as compared to the cryo-state. (Note that in the cryo-state the diameter of the composites was around 160-200 nm, whereas in the dried state it was found to be around 150-200 nm. This perhaps alludes to the stabilizing effect the molybdates exert on the vesicular structure and could be exploited for the construction of stable molybdate based materials.) Second, TEM images of the composite in the dried state (Figure 3b), instead of showing spherical objects with higher contrast along the edges (as would be expected for a shell), revealed just the opposite, that is, a high contrast material in the center in addition to that of the periphery. A similar picture is observed in the dark field mode (Figure 3c). EDX coupled with STEM confirms that contrast enhancing material to be molybdate (Figure 3d). The apparent anomalous presence of a higher contrast material (molybdate) in the core of the composites instead of only along the surface in the dried state could be explained as follows. The effect could be seen as a result of aggregation due to the drying of multiple DOTAP-molybdate vesicles leading to the distribution of molybdate all over the core and the periphery of the particles in the dried state. More precisely, the vesicular DOTAP structure
Figure 3. (a) Cryo-TEM image of the DOTAP-molybdate composite showing agglomeration of spheres. An individual sphere is also shown (with black arrow); note the higher contrast on the edges of the spheres. (b) TEM images of the composites in the dried state. Note the higher contrast inside due to the permeation of molybdates inside the vesicles in the composite as an effect of drying. (c) Darkfield TEM image of the composite together with an EDX analysis (d) along the gray line in (c), implying the presence of molybdate (with carbon) in the composite.
is not molybdate-impermeable.19 Upon drying, molybdates percolate from the surface to the vesicle interior, and this phenomenon could explain the above TEM observation. It is further worth mentioning that upon increasing temperature cylindrical composite species were observed, implying the possibility of changing the topology of the superstructure by changing the scaffold structure, at least in principle. We now use vibrational spectroscopy to expound further on our TEM results. The DOTAP dispersion in itself does not reveal significant Raman active modes in the fingerprint regime (700-1000 cm-1). It is known that in this regime molybdate (heptamolybdate) shows three characteristic bands at 785, 900 (νMo-O-Mo), and 945 (νMod -1 1 O) cm . Interestingly, in the composite, in addition to the above two Raman bands stemming from the symmetric stretch (νMo-O-Mo) of the molybdates (observed at 788, 896 cm-1), a significant red-shift in the third band (νModO) is further observed and the band appears at 954 cm-1, implying protonation of the ModO terminal bond, which in turn forms an H-bond with DOTAP (Figure 4). To confirm these observations, further FTIR spectroscopy was undertaken. FTIR spectra of the dried composite were in accordance with the results obtained from Raman spectroscopy. The spectra revealed in addition to a superposition of the IR bands of DOTAP and heptamolybdate the following further features. For instance, the ∂N-H band (1400 cm-1) of heptamolybdate is very weak (or almost absent) in the composite, implying substitution of ammonium by DOTAP cations (Figure 5). The red shifts of the bands in the composite characteristic of Mo-O-Mo and Mod O bonds (in the range of 840-920 cm-1) with respect to the starting molybdate most likely imply weakening of those bonds (19) Volkmer, D.; Bredenko¨tter, B.; Tellenbro¨ker, J.; Ko¨gerler, P.; Kurth, D. G.; Lehmann, P.; Schnablegger, H.; Schwahn, D.; Piepenbrink, M.; Krebs, B. J. Am. Chem. Soc. 2002, 124 (35), 10489-10496
Letters
Figure 4. Raman spectra of the DOTAP-molybdate composite and DOTAP.
Figure 5. FTIR spectra of the DOTAP-molybdate composite and heptamolybdate.
in the composite and could be a result of H-bonded interaction of the molybdate with the DOTAP scaffold. O‚‚‚O distances between the two CdO groups (6.652 Å) in DOTAP20 and those of the ModO bonds (6.646 Å)1 in the starting molybdate reinforce the steric compatibilty for such H-bond formation. More precisely, we put forward that a O-H‚‚‚OdC bond is formed between the OH of protonated molybdates and the OdC of DOTAP in the composite (Figure 1). This conclusion is based on the following observations: (1) from the protonation of ModO terminal bonds as indicated by Raman and FTIR spectroscopic results and (2) from the aforementioned matching O‚‚‚O distances of 6.6 Å between the terminal molybdates and the carbonyls of DOTAP. (20) Generosi, J.; Castellano, C.; Pozzi, D.; Congiu Castellano, A.; Felici, R.; Natali, F.; Fragneto, G. J. Appl. Phys. 2004, 96 (11), 6839-6844.
Langmuir, Vol. 24, No. 3, 2008 669
All these pieces of evidence indicate the formation of a POM supersphere on the DOTAP vesicular scaffold. Having shown the successful formation of the POM supersphere with embedded DOTAP, we now try to explain the plausible arrangement of the molybdates on the DOTAP vesicle surface from the results of our electrophoretic mobility experiments discussed before (Figure 2). These experiments show that the surface charge density of both the DOTAP vesicle and the composite (for M/D ≈ 3 and higher) is same but only has opposite signs (i.e., +5 µm cm V-1 s-1 in DOTAPs and -5 µm cm V-1 s-1 in the composite) (Figure 2). On the other hand, a DOTAP molecule carries a unit positive charge, whereas heptamolybdate has a charge of -6. From our above experimental results (i.e., charge inversion at M/D ≈ 3 and higher), it follows that in the composite for every three DOTAP molecules, there is only one heptamolybdate. This picture is further consistent with the surface area of DOTAP20 and heptamolybdate1 reported in the literature. So, our POM supersphere is most likely a vesicular DOTAP covered with a heptamolybdate monolayer where every three DOTAP molecules bind with one heptamolybdate of the monolayer. To conclude, we have here shown the successful use of a scaffold to synthesize unusually large polyoxomolybdate spheres in a controlled way. In contrast to the existing techniques, especially the ones introduced by the Liu group5-8 to self-assemble complex POMs to large surface vesicles in solution, our route relies on the scaffold and hence is independent of the complexity of the starting oxomolybdate building blocks. To prove this point, a simple anionic molybdate has been assembled to a supersphere in this letter. It further implies that more complex POMs could be assembled in a directed manner to yield complex mesoscopic POM aggregates. The option of further varying the overall topology of such POM aggregates just by changing the shape of the vesicle is an added bonus of this design, at least in principle. The method introduced here could lead to a new type of engineered mesoscopic polyoxomolybdate chemistry and can meet the need for a tailor-made large surface area POM based material. Acknowledgment. This research was financially supported by NWO-CW. LA703467D
