Supramolecular Self-Assembled Arrangements of Maltose Glyconanoparticles

5124

Langmuir 2008, 24, 5124-5128

Supramolecular Self-Assembled Arrangements of Maltose Glyconanoparticles Martina Fuss,† Mo´nica Luna,*,† David Alca´ntara,‡ Jesu´s M. de la Fuente,‡,§ Soledad Penade´s,‡ and Fernando Briones† Instituto de Microelectro´ nica de Madrid (CSIC), E-28760 Tres Cantos, and Laboratory of Glyconanotechnology, CIC biomaGUNE-CIBER-BBN and Instituto de InVestigaciones Quı´micas (CSIC), E-20009 San Sebastia´ n, Spain ReceiVed NoVember 28, 2007. In Final Form: January 21, 2008 Our group previously reported the preparation of water-soluble Au-FexOy nanoparticles functionalized with a maltose neoglycoconjugate. A fraction soluble in methanol was also separated and originated a new supramolecular polymeric aggregate. We report here the full characterization of this novel material by transmission electron microscopy (TEM), fluorescence emission, and atomic force microscopy. By means of noncontact dynamic atomic force microscopy, we have been able to obtain information about the organization of the organic components of the polymers, which eluded TEM analysis. We have observed that polymers packed in units about 65 nm in length and 40 nm in width on Au surfaces. The nanoparticles seem to be encapsulated by the organic material. We propose interactions between the sugar residues and the amphiphilic character of the maltose neoglycoconjugate (with a lipophilic undecane spacer) as responsible for the origin of these amazing supramolecular arrangements.

Introduction Since the field of glyconanotechnology was introduced in 2001,1 various kinds of nanostructures have been reported from isolated glyconanoparticles to self-assembled monolayers. These nanostructured biomaterials have demonstrated very interesting applications:2 gold glyconanoparticles have been used as multivalent systems to study and evaluate biological interactions where carbohydrates are involved3 and to isolate and characterize proteins.4 The extremely high stability and water solubility of these glyconanoparticles have revealed unusual physical properties5 and surprising new assemblies.6 As an example of these structures, we obtained a white residue, isolated during the preparation of water-soluble Au-FexOy glyconanoparticles functionalized with a maltose neoglycoconjugate.7 Analysis of this fraction, soluble in methanol, by transmission electron microscopy (TEM) showed a regular alignment (Figure 1) of gold nanoparticles, whereas no linear order was observed for the non-methanol-soluble residue. The concept of coassembling inorganic precursor molecules with amphiphile organic molecules for controlling the structure * To whom correspondence should be addressed. E-mail: mluna@ imm.cnm.csic.es. † Instituto de Microelectro ´ nica de Madrid. ‡ CIC biomaGUNE-CIBER-BBN and Instituto de Investigaciones Quı´micas. § Current address: Instituto de Nanociencia de Arago ´ n, University of Zaragoza, E-50009 Zaragoza, Spain. (1) de la Fuente, J. M.; Barrientos, A. G.; Rojas, T. C.; Rojo, J.; Can˜ada, J.; Ferna´ndez, A.; Penade´s, S. Angew. Chem., Int. Ed. 2001, 40, 2257-2261. (2) de la Fuente, J. M.; Penades, S. Biochim. Biophys. Acta 2006, 1760, 636651. (3) de la Fuente, J. M.; Penade´s, S. Glycoconjugate J. 2004, 21, 149-163. (4) Chen, Y.-J.; Chen, S.-H.; Chien, Y.-Y.; Chang, Y.-W.; Liao, H.-K.; Chang, C.-Y.; Jan, M.-D.; Wang, K.-T.; Lin, C.-C. ChemBioChem 2005, 6, 1169-1173. (5) Crespo, P.; Litra´n, R.; Rojas, T. C.; Multigner, M.; de la Fuente, J. M.; Sa´nchez-Lo´pez, J. C.; Garcı´a, M. A.; Hernando, A.; Penade´s, S.; Ferna´ndez, A. Phys. ReV. Lett. 2004, 93, 087204. (6) Rojas, T. C.; de la Fuente, J. M.; Barrientos, A. G.; Penade´s, S.; Ponsonnet, L.; Ferna´ndez, A. AdV. Mater. 2002, 14, 585-588. (7) de la Fuente, J. M.; Alca´ntara, D.; Eaton, P.; Crespo, P.; Rojas, T. C.; Ferna´ndez, A.; Hernando, A.; Penade´s, S. J. Phys. Chem. B 2006, 110, 1302113028.

Figure 1. Transmission electron micrograph of malto-Au-FexOy(polymer). The nanoparticles appear orderly aligned. Two domains have been highlighted, with arrows indicating the respective direction of parallel alignment. The inset shows the particles size distribution.

and organization of inorganic materials has successfully been used to produce materials with both scientific and technological importance. Different methods have been used to prepare gold nanoparticles with supramolecular structures8 or with subnanometer-ordered domains in the ligand shell.9 Block copolymers10 or functionalized block comicelles11 are able to embed gold nanoparticles in a controlled way. Commercially available surfactants drive the self-organization of soluble Zintl clusters,12 and peptide-amphiphile nanofibers control the morphology of CdS nanocrystals.13 These organization processes are not nanoparticle or metal specific but depend mainly on the nature, (8) Kim, J.-U.; Cha, S.-H.; Shin, K.; Jho, J. Y.; Lee, J. Ch. J. Am. Chem. Soc. 2005, 127, 9962-9963. (9) Jackson, A. M.; Myerson, J. W.; Stellacci, F. Nat. Mater. 2004, 3, 330336. (10) Chiu, J. J.; Kim, B. J.; Kramer, E. J.; Pine, D. J. J. Am. Chem. Soc. 2005, 127, 5036-5037. (11) Wang, H.; Lin, W.; Fritz, K. P.; Scholes, G. D.; Winnik, M. A.; Manners, I. J. Am. Chem. Soc. 2007, 129, 12924-12925. (12) Sun, D.; Riley, A. E.; Cadby, A. J.; Richman, E. K.; Korlann, S. D.; Tolbert S. H. Nature 2006, 44, 1126-1130. (13) Sone, E. D.; Stupp, S. I. J. Am. Chem. Soc. 2004, 126, 12756-12757.

10.1021/la703716g CCC: $40.75 © 2008 American Chemical Society Published on Web 04/11/2008

Arrangements of Maltose Glyconanoparticles

Langmuir, Vol. 24, No. 9, 2008 5125

concentration, and ratio of the organic ligands and on the experimental conditions of the methods used. The question is whether the concept can be generalized to all organic systems. Amphiphile carbohydrate molecules (glycolipids, neoglycolipids, or polysaccharides) are ideal candidates because they tend to form supramolecular structures in water.14 The amphiphilic nature of the maltose neoglycoconjugate 11,11′-dithiobis[undecanyl-β-maltoside] (1) used in the preparation of the glyconanoparticles may be the origin of the supramolecular assemblies observed for the methanol-soluble white residue fraction isolated in the preparation of the Au-FexOy glyconanoparticles. In this work we also characterize both glyconanoparticle fractions by fluorescence emission. Metals are usually considered nonluminescent; however, visible luminescence has been measured from metal colloidal nanoparticles. Chemiluminescence in the agglomeration of metal clusters (Cu, Ag, Au) was for the first time observed by Ertl and colleagues.15 In recent years there has been marked progress in studying the luminescence properties of metal nanoparticles.16-21 Wilcoxon et al.16 found quantum efficiencies of 10-4-10-5 for nanoparticles 5 nm in size, but 15 nm nanoparticles did not luminesce. The result where smaller clusters lead to lower emission wavelengths (higher energies)18,20 and higher quantum yields20 has been corroborated by some studies. Huang et al.18 showed that the real improvement in emission efficiency is originated by clusters less than 2 nm in diameter. They also found that luminescence efficiency varied with the monolayer ligand. It is generally accepted that the emission enhancement of small Au clusters occurs as a consequence of the discretization and gradual separation of the electronic levels as size decreases, which facilitates radiative emissions. In this work, by means of noncontact dynamic atomic force microscopy (AFM) we have also been able to obtain information about the topography of the organic coating of these periodically arranged gold nanoparticles, which eluded TEM analysis. In the AFM technique, contact operation modes are most widely utilized. Intermittent contact, also known as “tapping”, is a method where the tip briefly contacts the surface during each oscillation cycle. This method has advantages over the standard contact mode for the fact that it does not produce lateral friction forces on the observed object which could cause its displacement. Also, because relatively stiff tips are used (force constants in the 50 N/m range) in combination with large oscillation amplitudes, the adhesive forces can be easily overcome. Nevertheless, intermittent contact modes operated in air are not adequate for the study of delicate adsorbents, including weakly bound molecules or individual molecular polymers which could be easily detached or deformed by the contact forces. Therefore, for the present study of maltose-protected nanoparticles, we employed a dynamic noncontact AFM mode based on the change in resonance oscillation amplitude of the cantilever

by the effect of attractive van der Waals forces.22-24 The oscillation amplitude decreases as the tip approaches the surface (without touching it), similarly to the intermittent contact mode. This amplitude is the signal used as the Z-piezo feedback to obtain a topographic image. The main difference is that, in this noncontact mode, the tip never makes adhesive or repulsive contact with the surface. Parameters such as the cantilever driving oscillation amplitude, force constant, resonance frequency, and oscillation amplitude set point are optimized for the operation out of contact.25 In the noncontact mode, images are taken at a typical distance of about 1-5 nm from the surface.26,27 The peak-to-peak oscillation amplitude for noncontact operation ranges from 0.5 to 10 nm for the cantilevers used, whichswith about 1 N/msare somewhat stiffer than those typically used for the contact mode, but less than for the intermittent contact mode. In particular, this method, where no voltage is required between the tip and sample, has been extensively used to image the topography of samples as delicate as liquid droplets25 and films.28 One important characteristic of this mode is that the change in phase contrast within the image is very small compared to that occurring in intermittent contact mode29 because energy dissipation is also much smaller. This is an important signature that permits in situ monitoring of the noncontact condition during imaging. Moreover, although the changes in energy dissipation are small, they are still a source of chemical contrast that allows distinguishing between different materials as reported also for intermittent contact.30

(14) For recent reports see: John, G.; Masuda, M.; Okada, Y.; Yase, K.; Shimizu, T. AdV. Mater. 2001, 13, 715-718. Jung, J. H.; John, G.; Yoshida, K.; Shimizu, T. J. Am. Chem. Soc. 2002, 124, 10674-10675. Numata, M.; Asai, M.; Kaneko, K.; Bae, A.-H.; Hasegawa, T.; Sakurai, K.; Shinkai, S. J. Am. Chem. Soc. 2005, 127, 5875-5884. (15) Ko¨nig, L.; Rabin, I.; Schulze, W.; Ertl, G. Science 1996, 274, 13531355. (16) Wilcoxon, J. P.; Martin, J. E.; Parsapour, F.; Wiedenman, B.; Kelley, D. F. J. Chem. Phys. 1998, 108, 9137-9143. (17) Mohamed, M. B.; Volkov, V.; Link, S.; El-Sayed, M. A. Chem. Phys. Lett. 2000, 317, 517-523. (18) Huang, T.; Murray, R. W. J. Phys. Chem. B 2001, 105, 12498-12502. (19) Zheng, J.; Petty, J. T.; Dickson, R. M. J. Am. Chem. Soc. 2003, 125, 7780-7781. (20) Zheng, J.; Zhang, C.; Dickson, R. M. Phys. ReV. Lett. 2004, 077402. (21) Hodes, G. AdV. Mater. 2007, 19, 639-655.

(22) Anczykowski, B.; Kru¨ger, D.; Babcock, K. L.; Fuchs, H. Ultramicroscopy 1996, 66, 251-259. (23) Ku¨hle, A.; Sørensen, A. H.; Bohr, J. J. Appl. Phys. 1997, 81, 6562-6569. (24) Luna, M.; Colchero, J.; Baro´, A. M. J. Phys. Chem. B 1999, 103, 95769581. (25) Luna, M.; Colchero, J.; Go´mez-Herrero, J.; Baro´, A. M. Appl. Surf. Sci. 2000, 157, 285-289. (26) Luna, M.; Colchero, J.; Baro´, A. M. Appl. Phys. Lett. 1998, 72, 34613463. (27) de Pablo, P. J.; Colchero, J.; Luna, M.; Go´mez-Herrero, J.; Baro´, A. M. Phys. ReV. B 2000, 61, 14179-14183. (28) Gil, A.; Colchero, J.; Luna, M.; Go´mez-Herrero, J.; Baro´, A. M. Langmuir 2000, 16, 5086-5092. (29) Haugstad, G.; Jones, R. R. Ultramicroscopy 1999, 76, 77-86. (30) Cleveland, J. P.; Anczykowski, B.; Schmid, A. E.; Elings, V. B. Appl. Phys. Lett. 1998, 72, 2613-2615.

Methods Preparation of Maltose Gold-Iron Oxide Glyconanoparticles malto-Au-FexOy and malto-Au-FexOy(polymer). The synthesis, isolation, and characterization of gold-iron oxide glyconanoparticles were previously reported.7 In brief, to a solution of the disulfide 1 (100 mg, 0.094 mmol, 3 equiv) in methanol (14 mL) was added a solution of FeCl3 (2.5 mg, 0.01 mmol, 0.25 equiv) in water (0.63 mL) followed by the addition of a solution of HAuCl4 (21 mg, 0.04 mmol, 1 equiv) in water (2.5 mL). The mixture was left for 5 min at 60 °C, and then an aqueous solution of NaBH4 (1 M, 65 mg, 1.7 mmol, 27.5 equiv) was added in small portions with rapid stirring. The brown suspension formed was stirred for an additional 2 h. The crude product was separated by decantation from the methanolic solution. The brown residue was dispersed in NANOpure water (∼15 mL), loaded into a centrifugal filter device (Centriplus YM30, Microcon, MWCO ) 30 000), and subjected to centrifugation (3000g, 40 min). The dark glyconanoparticle residue corresponds to maltoAu-FexOy nanoparticles. On the other hand, the colorless methanolic dispersion obtained from the reaction crude was evaporated, and a white residue was isolated. The residue was dispersed in water, centrifugally filtered (Centriplus YM30, Microcon, MWCO ) 30 000, 3000g, 40 min), and washed with water (8 × 5 mL). The residue was dissolved in water and lyophilized to give malto-Au-FexOy(polymer) as a white solid. Average diameter and number of gold atoms: 1.5 nm and 79. 1H NMR (500 MHz, D O): δ (ppm) 5.32 (s, 1H, H-1′), 4.37 (s, 1H, 2

5126 Langmuir, Vol. 24, No. 9, 2008 H-1), 4.00-3.30 (m, 13H), 2.70 (s, 2H, CH2S), 1.85-1.20 (m, 17H). UV (H2O, 2 mg‚mL-1): λ ) 500 nm, broad band. Fluorescence (H2O): λex ) 254 nm, λem ) 599 nm. ICP: 0.815% Au, 0.039% Fe. Anal. Calcd for (C23H44O11S)2050Au67Fe12: C, 51.7; H, 8.1; S, 6.0. Found: C, 50.6; H, 7.9; S, 5.8. TEM Characterization. Drops (10 µL) of the aqueous solutions (0.1 mg/mL) of the glyconanoparticles were placed onto a copper grid coated with a carbon film. The grid was left to dry in air for several hours at room temperature. TEM examination of the samples was carried out at 200 keV with a Philips CM200 microscope. Sample Preparation for AFM Analysis. A droplet (10 µL) of a solution of maltose-protected AuFe nanoparticles (70 µg/mL) was deposited onto a metallic substrate e-beam evaporated with Cr (10 nm) and gold (100 nm). These samples were then left to dry at about 6 °C for 24 h, allowing a slow evaporation of the water. AFM imaging took place at normal ambient conditions (23 °C and 35% RH). Atomic Force Microscope Operation. A commercial atomic force microscope (Nanotec Electro´nica S.L., Spain) was operated in noncontact mode. The cantilevers used (Olympus Optical LDT, Japan) had a nominal force constant of 0.73 N/m and a resonance frequency of 71 kHz. The peak-to-peak oscillation amplitude applied during imaging was about 4 nm.

Fuss et al.

Figure 2. Transmission electron micrograph of malto-Au-FexOy nanoparticles.

Results and Discussion TEM and Fluorescence Characterization of Maltose GoldIron Oxide Polymer Glyconanoparticles. In a previous paper, we reported the synthesis of magnetic Au-FexOy nanoparticles functionalized with the disaccharides maltose (malto) and lactose (lacto) covalently linked to the inorganic cluster by an aliphatic spacer ending in a thiol group.7 This approach opened the way for tailoring glyconanoparticles with a variety of carbohydrate ligands and obtaining very useful magnetic properties for future applications as ferrofluids or for molecular imaging and tumor targeting.31 The so-prepared glyconanoparticles have a metallic core diameter of less than 2 nm. They form stable dispersions, and even after several months, flocculation or aggregation is not observed. The preparation of the nanoparticles was carried out in a methanolic solution of FeCl3 and HAuCl4 (ratio 1:4). The salt solution was reduced with NaBH4 at 60 °C in the presence of disulfide derivatives of the sugar conjugates to give a stable dispersion of the nanoparticles. In this experiment, the magnetic Au-FexOy nanoparticles were capped with the neoglycoconjugate 11,11′-dithiobis[undecanylβ-maltoside] (1). After the nanoparticles were isolated from the methanolic solution by decantation, we could obtain, after evaporation, a white residue which was also soluble in water. This white residue was only observed when 11,11′-dithiobis[undecanyl β-maltoside] was used. Preparation using a lactose neoglycoconjugate did not show this behavior. The proton NMR of this white residue gave the same spectrum as the free maltoside disulfide 1, suggesting to us that this residue did not contain gold nanoparticles. However, examining this fraction by TEM once dried on an amorphous carbon support, we observed7 nanoparticles well ordered in a kind of polymeric parallel arrangement (Figure 1). The decanted fraction provides TEM images (Figure 2) with randomly distributed nanoparticles. Thus, two different types of structures were isolated: one where the nanoparticles are aligned (malto-Au-FexOy(polymer), Figure 1) and another with nonordered nanoparticles (malto-Au-FexOy, Figure 2). Although care was taken to work with minimum beam fluency, we could not exclude the possibility that the nanoparticles observed by TEM in the polymeric fraction formed as a (31) Berry, C. C.; Curtis, A. S. G. J. Phys. D: Appl. Phys. 2003, 36, R198R206.

Figure 3. Fluorescence emission spectra of malto-Au-FexOy(polymer), continuous line, and malto-Au-FexOy, dotted line. λex ) 254 nm. The malto-Au-FexOy(polymer) shows fluorescence at 600 nm. The higher intensity peaks at 508 nm are satellites of the irradiation wavelength (254 nm).

consequence of TEM electron irradiation of the original molecular gold polymer.8 Fluorescence spectra were measured for both fractions. The fluorescence spectrum showed an emission band at 600 nm when the malto-Au-FexOy(polymer) sample was UV irradiated at 254 nm (Figure 3, continuous line). This fluorescence emission was not observed in the case of the nonordered malto-Au-FexOy glyconanoparticle fraction (Figure 3, dotted line). Since the nanoparticles in the polymeric form malto-Au-FexOy(polymer) present a smaller diameter (inset Figure 1, mean value of the diameter 1.58 ( 0.03 nm) than those forming the nonordered part (malto-Au-FexOy, mean value of the diameter 2.2 ( 0.3 nm),7 fluorescence emission in the case of malto-Au-FexOy(polymer) but not of the other fraction might be due to this size difference as previously observed.16,21 The larger nanoclusters may not exhibit discrete or sufficiently separated energy levels and may be therefore unable to produce radiative recombinations with an emission efficiency comparable to that of the smaller ones. TEM micrographs of the malto-Au-FexOy(polymer) fraction show very small Au-FexOy nanoparticles perfectly aligned and probably encapsulated in organic material (Figure 1). Compositional analysis of this polymer gives 2050 maltose molecules per nanoparticle of Au-FexOy, although the dialysis procedure used to prepare the sample eliminates single maltose neoglycoconjugates. Therefore, both TEM and chemical analysis indicate a self-assembled organization of the neoglycoconjugates during the nanoparticle synthesis. The amphiphilic character of the maltose neoglycoconjugate, with a hydrophilic head and a

Arrangements of Maltose Glyconanoparticles

Figure 4. Cartoon showing the micelle model for malto-Au-FexOy(polymer) and the corresponding approximate dimensions superposed on a TEM micrograph.

lipophilic spacer, may be the origin of the self-assembled arrangements in water solution. As a consequence, rows of aligned gold nanoparticles appear encapsulated by maltose neoglycoconjugate molecules forming a kind of nanotube of organic material. Thus, maltose neoglycoconjugate molecules selfassemble in a tubular, micellelike arrangement containing AuFexOy nanoparticles inside the tubes. Figure 4 shows a cartoon of this micellar model superposed to a TEM micrograph of the sample. It must be highlighted that the distance between two consecutive gold nanoparticle rows is around 3.4 nm. This distance corresponds very well to half of the expected diameter of the micelle nanotubes, according to the known length of the maltose molecular chains (2.5 nm). Therefore, a model in which nanotubes are packed in various layers could explain the distances between rows found on TEM images (Figure 4). The dark contrast in the TEM image corresponds to the projection on the image plane of the nanoparticles superposed from various piled micelle nanotube layers. Results on encapsulation of gold nanoparticles in a similar kind of polymer have been recently published.32 Neoglycolipids are known to form micelles and other types of aggregates in water.14 The intermolecular forces involved are of diverse nature (van der Waals, hydrophobic, hydrogen bond interactions), but water desolvation and the stereochemistry of the carbohydrate head confer the stability and specificity to the interactions.33-35 Thus, depending on the sugar configuration of the monosaccharide (glucose, galactose, mannose, fucose, etc.) components of the more complex sugars (di-, tri-, and tetraoligosaccharides, etc.) the interaction established may be rather different. We have also prepared glyconanoparticles with 11,11′-dithiobis[undecanyl-β-lactoside],1,36 but we never observed the formation of lacto-Au(polymer). Probably, the ability of the maltose neoglycoconjugates to form the tubular micellelike aggregates is due to additional carbohydrate-carbohydrate interactions37 between the maltose sugar residues as described in the crystal structure of the maltose disaccharide.38 (32) Yang, B.; Kamiya, S.; Yoshida, K.; Shimizu, T. Chem. Commun. 2004, 500-501. (33) Lemieux, R. U. Acc. Chem. Res. 1996, 29, 373-380. (34) Israelachvili, J.; Wennerstro¨m, H. Nature 1996, 379, 219-225. (35) Cheng, Y-K.; Rossky, P. J. Nature 1998, 392, 696-699. (36) Barrientos, A. G.; de la Fuente, J. M.; Rojas, T. C.; Ferna´ndez, A.; Penade´s, S. Chem.sEur. J. 2003, 9, 1909-1921. (37) Rojo, J.; Morales, J. C.; Penade´s, S. Top. Curr. Chem. 2002, 218, 45-92. (38) Takusagawa, F.; Jacobson, R. A. Acta Crystallogr. 1978, B34, 213-218.

Langmuir, Vol. 24, No. 9, 2008 5127

Figure 5. Noncontact AFM topographical image (500 × 446 nm) of polymeric structures of malto-Au-FexOy(polymer) observed in the center of the image on a polycrystalline Au surface. The image shown is represented using the three-dimensional mode to highlight the presence of two elongated structures (areas A and B).

From ICP analysis and the measured diameter of the metallic clusters, we obtain an approximate gold and iron atomic content of Au67Fe12Ox. EDX analysis corroborates that atomic ratio. However, the organic part determination gives a content of 2050 maltose molecules per cluster. This implies that not all the maltose chains are attached to the metallic nanoclusters containing only 67 Au atoms. There are not enough binding sites available on the cluster, even assuming that all the gold is at the cluster surface. This fact becomes obvious in the TEM image, where the metal density seems much higher than expected. The metallic clusters appear to be distributed inside the organic micellar nanotubes with a periodicity which is difficult to determine on the TEM images due to the unknown number of superposed micellar layers on the microscope preparation. Noncontact AFM Characterization. Preliminary AFM analysis of malto-Au-FexOy(polymer) on mica confirmed this encapsulation.7 We now explore the structure of these micellar nanotubes, deposited on a gold thin film surface by noncontact AFM techniques. The deposition of a droplet of malto-Au-FexOy(polymer) solution resulted in the observation of elongated structures which were often observed in groups of two, as in the example of Figure 5, where we have named the two structures with the letters A and B to identify them. Less commonly, these structures were found forming more complex formations such as the one shown in Figure 6a,b. In Figure 6c,d we have circled some of these elongated structures to help the eye to differentiate them from the substrate roughness, and we have named them with the letters C, D, E, F, G, and H. We performed a statistical analysis of the topography of more than 30 elongated structures to obtain an averaged value of their dimensions. Laterally, more than 80% of the structures analyzed measured 65 ( 5 nm in length and 40 ( 5 nm in width. The average height value of the structures is 12 ( 3 nm. The uncertainty value of (3 nm found for the height measurement is related to the fact that the structures are not adsorbed onto a flat surface. A roughness analysis of the gold grain surface gives a Gaussian distribution for the height histogram. The Gaussian curve is peaked at 2 nm, and the data extend over 4 nm. In Figure

5128 Langmuir, Vol. 24, No. 9, 2008

Fuss et al.

association of micelle nanotubes with metallic nanoclusters inside each nanotube. In fact, we believe that the elongated structures found correspond to the domains observed by TEM within which the nanoparticles conserved their parallel alignment. According to this model, the height measured experimentally corresponds to two layers of the tubular micelles. From the TEM results we can conclude that it is common to have more than one of these layers (the separation between rows in Figure 4 is 3.4 nm instead of 7 nm). Parts a and b of Figure 6 represent the topography and the phase images taken simultaneously at the minimum tip-sample interaction. In the image of Figure 6b a small phase contrast between the organic material and the gold grains can be observed (areas formed by organic material show a darker color). This result indicates that although the difference in energy dissipation between both materials is small in this operation mode, it represents a source of chemical contrast. This additional information is especially important in distinguishing between structures with similar topography, as happens in this case where the elongated structures present dimensions similar to those of the gold grains.

Conclusion

Figure 6. Topographic (a, c) and phase (b, d) images of a complex structure found on gold surfaces. (c) and (d) are zoomed-in zones of (a) and (b), respectively, marked with a rectangle in each image. In particular, the dimensions of the elongated structure D are estimated to be 40 nm for the width (from profile 1, panel e), 32 nm for the semilength (from profile 2, panel f), and 11 nm for the height. A statistical analysis of more than 30 structures gave average dimensions of 40 ( 5 nm × 65 ( 5 nm. These dimensions are similar to those of the ordered domains observed in TEM images.

6e,f we show the dimensions of a particular elongated structure (D). Profile 1 (Figure 6c) corresponds to the width of the elongated structure graphed in Figure 6e, and profile 2 (Figure 6c) corresponds to the semilength of structure D, which is shown in Figure 6f. It is interesting to observe that the lateral dimensions of the elongated structures are in good agreement with the dimensions of the domains revealed by TEM analysis (Figure 1). If we apply the above proposed model to this polymeric material adsorbed on gold, the elongated structures (like areas A, B, C, D, E, and F in Figures 5 and 6) can be interpreted as a set of cylindrical micelles consisting of a core of aligned gold nanoclusters and an envelope of maltose neoglycoconjugate (Figure 4). In other words, the elongated structures would be formed by the parallel

In this work we report, for the first time, the full characterization of maltose glyconanoparticle polymeric nanostructures by TEM, noncontact AFM, and fluorescence emission. We also propose a model to explain this new supramolecular polymeric structure based on carbohydrate-carbohydrate interactions between the sugar residues and the amphiphilic character of the maltose neoglycoconjugate, with a hydrophilic head and a lipophilic spacer. The orderly aligned small Au-FexOy nanoclusters would be embedded inside organic nanotubes made of maltose neoglycoconjugates as a kind of a micellar arrangement. These maltose glyconanoparticle conjugates adopt specific arrangements when adsorbed on gold. By means of noncontact dynamic AFM, we obtained experimental information about the organic coverage of this self-assembled arrangement. We observe the formation of elongated structures that present typical dimensions of 65 nm length and 40 nm width. We interpret these structures as aggregations of the tubular micelles. The measured height of the subunits corresponds within the model to two layers of piled nanotubes. The metallic nanoclusters found forming the malto-Au-FexOy(polymer) structures show a clear fluorescence emission band at 600 nm. Acknowledgment. This work was supported by the Spanish Ministry of Education and Science through Grants CTQ200507993-C02-02/BQU, CTQ2005-07993-C02-01/BQU, NAN200409125-C07-01, and NAN2004-09125-C07-02. M.F. acknowledges financial support from the Spanish Ministry of Industry through the project Biosense FIT-010000-2006-98. J.M.d.l.F. and M.L. acknowledge financial support from the Ministry of Education and Science through the Ramon & Cajal program. LA703716G