WO2008044896A1 - Carbon nanotube-dendron composite and biosensor comprising the same - Google Patents

Abstract

The present invention relates to a carbon nanotube (CNT)-Dendron composite and a biosensor for detecting a biomolecule comprising the CNT-Dendron composite. The noncovalent functionalization of CNTs by a Dendron 3an be a new approach toward sensible nanobio-devices, not only by introducing biomolecular probes on CNTs without disruption of the electronic network of the tubes, but also by providing the immobilized probe molecules with an ample space enough to minimize steric hindrance for the unhindered interaction with their target species.

Description

CARBON NANOTUBE-DENDRON COMPOSITE AND BIOSENSOR COMPRISING THE SAME

CROSS REFERENCE TO RELATED APPLICATION

The present application claims priority to and the benefit of United States provisional application No. 60/829,262 filed in the United State Patent and Trademark Office on October 12, 2006, the entire content of which is incorporated hereinto by reference.

BACKGROUND OF THE INVENTION

(a) Field of the Invention

The present invention relates to a carbon nanotube (CNT)-Dendron composite and a biosensor for detecting a biomolecule comprising the CNT-Dendron composite. More specifically, the present invention relates to CNT-Dendron composite where a plurality of termini of a branched region of at least a Dendron comprising a branched region and a linear region are bound non-covalently to the sidewall of CNT. The noncovalent functionalization of CNT by a Dendron is a new approach toward sensible nanobio-devices, not only by introducing biomolecular probes on CNTs without disruption of the electronic network of the tubes, but also by providing the immobilized probe molecules with an ample space enough to minimize steric hindrance for the unhindered interaction with their target species.

(b) Description of the Related Art

Intelligent integration of biomolecules with nanoscale materials such as CNTs is one of the key processes toward the realization of successful nanobiosystems, including biosensors and drug delivery systems. CNTs, comprising multiple concentric shells and termed multi-wall carbon nanotubes (MWNTs), were discovered by Iijima in 1991. Subsequent to this discovery, single-wall carbon nanotubes (SWNTs), comprising a single graphene rolled up on itself, were synthesized in an arc-discharge process using carbon electrodes doped with transition metals. These carbon nanotubes (especially SWNTs) posses unique mechanical, electrical, thermal and optical properties, and such properties make them attractive for a wide variety of applications.

Several methods of preparing nanotubes have been described, including that by T. W. Ebbesen et al. Other methods for the chemical functionalization of nanotubes have also been described; there may be mentioned for example TSANG S. C. et al., Journal of the Chemical Society, Chemical Communications, Inorganica Chimica Acta,

However, they involve chemical reactions which either dramatically modify the geometry of the nanotubes (opening of the ends, partial destruction of the outer sheets), or destroy the intrinsic physical properties of the nanotubes. Nanotubes modified by such destructive methods are not therefore suitable for adsorption and/or immobilization at their outer surface of synthetic products or of biological macromolecules.

Depending on the technique and the conditions used, several structures of nanotubes may be prepared: the nanotubes have in particular so-called MWNT or single- SWNT. They can be completely, partially or not at all oxidized. When immobilizing probe biomolecules on nanomaterials, high density packing might seem to be desirable to achieve enhanced binding probability and capacity. Yet, it should also be considered to secure sufficient spacing between individual probes, because reserving enough room for each probe would reduce steric crowding and keep their pristine structure and biological activity, thereby ensuring unhindered interactions with their target species (Besteman, K.; Lee, J. O.; Wiertz, F. G. M.; Heering, H. A.; Dekker, C. Nano Lett. 2003, 3, 727-730), almost no considerations on such aspects have been made. Jiang et al suggested in their report that changing the reaction conditions for surface oxidation would control the immobilization density, but the control of spacing seems to be challenging yet.

SUMMARY OF THE INVENTION

The present inventors can provide the noncovalent functionalization of CNTs with selected dendrons. In addition, they found that the "fingertips" played a critical role in the binding efficiency. Moreover, since the dendrons on CNTs generate a certain distance between adjacent functionalities, further attached biomolecules at the apex are given proper space without unwanted aggregation. Hence, the Dendron lpproach can be utilized for enhancing the performance of various CNT-based nanobiosystems.

An object of the invention is to provide a CNT-Dendron composite where a plurality of termini of a branched region of at least a Dendron comprising a branched region and a linear region are bound non-covalently to the sidewall of CNT.

Another object of the present invention is to provide a biosensor for detecting a biomolecule comprising a CNT-Dendron composite where a plurality of termini of a branched region of at least a Dendron comprising a branched region and a linear region are bound non-covalently to the sidewall of CNT.

BRIEF DESCRIPTION OF THE DRAWINGS For a more complete understanding of the present invention, and the advantages thereof, reference is now made to the following descriptions taken in conjunction with the accompanying drawings, in which:

Fig. 1 shows three types of dendrons used for Example 1 in (a), and is a schematic illustration of the immobilization of gold nanoparticles and streptavidin on CNT surfaces through the fingertip-guided functionalization by dendrons (b).

Fig. 2 is an AFM image of a CNT functional ized by Dendron 1 , to which gold nanoparticles are attached, with the height values of three representative parts in (a), section profiles (b), and (c) is a corresponding model of the parts designated in (a), In Fig. 2, the height values without brackets in (a) indicate the heights from the bottom of the substrate, and the values with brackets are the differences from the point A.

Fig. 3 shows (a) visual comparison of CNTs without (left) and with (right) the Dendron 1 after sonication and centrifugation in DMF, and (b) Raman spectra of CNTs before (left) and after (right) the dendron treatment.

Fig. 4 is AFM images of CNTs functionalized by dendrons, to which gold nanoparticles are applied, showing discriminative binding efficiency of Dendrons 1, 2 and 3. CNTs were treated with (a) Dendron 1, (b) Dendron 2, and (c) Dendron 3, respectively.

Fig. 5 is spacing realized by the dendrons. (a) An example of small spacing between nanoparticles (~6 nm) for Dendron 1. The distance was measured to be the horizontal distance between the highest points of two NPs, indicated by arrows, (b)

Distribution of the spacing between adjacent nanoparticles for Dendron 1 (top) and Dendron 2 (bottom), (c) Streptavidins on a CNT modified by Dendron 1.

Fig. 6 shows tapping mode AFM images of a CNT obtained before (top) and after (bottom) the treatment of Dendron 1.

Fig. 7 shows AFM images obtained before and after applying gold nanoparticles to (a) CNTs functionalized with the Dendron 1 , (b) bare CNTs, and (c) Tween20-treated CNTs. There is no observable binding in cases of (b) and (c).

Fig. 8 represents control experiments with streptavidin on (a) bare CNTs and (b) Tween20-coated CNTs. (a) The image shows considerable nonspecific binding of streptavidin onto CNTs. (b) The image reveals that Tween20 repels streptavidin.

Fig. 9 is (a) t-Boc protected Dendron 1 analog (Dendron 1 '). (b) The Dendron 1 ' showed a similar binding affinity toward the sidewalls of CNTs.

DETAILED DESCRITPION OF THE PREFERRED EMBODIMENTS These and other objects of the invention will be more fully understood from the following description of the invention, the referenced drawings attached hereto and the claims appended hereto.

In the present application, "a" and "an" are used to refer to both single and a plurality of objects.

The term "dendrimer" is characterized by a core, at least one interior branched layer, and a surface branched layer (see Petar et al, Pages 641-645, In Chem. in Britain, August 1994). A "dendron" is a species of dendrimer having branches emanating from a focal point, which is or can be joined to a core, either directly or through a linking moiety to form a dendrimer. Many dendrimers include two or more dendrons joined to a common core. However, the term "dendrimer" may be used broadly to encompass a single dendron.

As used herein, "branched" as it is used to describe a macromolecule or a dendron structure is meant to refer to a plurality of polymers having a plurality of termini which are able to bind covalently or ionically to a substrate. In one embodiment, the macromolecule containing the branched structure is "pre-made" and is then attached to a substrate.

As used herein, "regular intervals" refers to the spacing between the tips of the size-controlled macromolecules, which is a distance from about 1 nm to about 100 nm so as to allow room for interaction between the target-specific ligand and the target substantially without steric hindrance. Thus, the layer of macromolecules on a substrate is not too dense for specific molecular interactions to occur.

In the present invention, the CNT can be prepared: the nanotubes have in particular so-called MWNT or SWNT. They can be completely, partially or not at all oxidized.

The present inventors provide the noncovalent functional ization of the sidewalls of CNTs by dendrons.

In an embodiment of the present invention, it is provided that a CNT-Dendron composite where a plurality of termini of a branched region of at least a Dendron comprising a branched region and a linear region are bound non-covalently to the sidewall of CNT.

In a preferred embodiment, the CNT-Dendron composite is a chemical compound having the following chemical structures. Chemical formula 1

Wherein, Rl is OH, or ^ ^V_./

R2 is a protecting group of tert-butyloxycarbonyl (t-BOC) or

The examples of the dendrons includes a compound of chemical formula I wherein Rl is OH and R2 is anthracene group, a compound of chemical formula wherein Rl is OH and R2 is t-BOC, a compound of chemical formula 1 wherein Rl is and R2 is anthracene group, and a compound of Chemical formula 1

-N

wherein the Rl is and R2 is t-BOC.

The CNT-Dendron composite, wherein the dendrons are spaced at regular intervals between about 0.1 nm and about 100 nm, preferably IOnm and 30 nm, and more preferably 10 nm and 25 nm among the linear regions. The the CNT of the composite maintains its π-configuration without disruption of the electronic network of the CNTs.

A dendron, which denotes a subunit of a dendrimer, inherits the superior properties of the macromolecule, such as monodispersity, a well-defined structure, and an easy tunability in their functionalities(Newkome₅ G. R.; Moorefield, C. N.; Vδgtle, F.

Dendritic Molecules: Concepts, Syntheses, Perspectives] VCH: Weinheim, Germany,

1996; Bosnian, A. W.; Janssen, H. M.; Meijer, E. W. Chem. Rev. 1999, PP, 1665-

1688; Grayson, S. M.; Frechet, J. M. J. Chem. Rev. 2001, 101, 3819-3868). A uniqueness of the dendrons, compared to the dendrimers, comes from their anisotropic shapes. By binding to a surface through its terminal groups, each dendron occupies its own space and thus the surface-bound dendrons can naturally secure a certain distance between their apex functionalities.

Dendrons have a unique anisotropic shape and an orthogonal functional group at their apex, and thus can generate a certain spacing between the functional groups upon the immobilization on surfaces. Atomic force microscope (AFM) imaging, dispersion experiments, and micro-Raman spectroscopy were employed for the characterization of the functional ization. The binding was found to be governed by the chemical nature of the terminal groups, namely, the "fingertips", through the comparison study on the adsorption efficiency of the Dendron lnalogs. Functional groups such as carboxylic acid group and benzyl amide group were effective for the cooperative binding.

To date, a few attempts to functionalize the sidewalls of CNTs with dendrons or dendrimers using covalent chemistry have been reported, such as the work to solubilize CNTs in organic solvents and in water, the work to introduce high-density functional groups around tubes, and the work for templated synthesis of quantum dots on CNT surfaces.²⁸ Little is known, however, about the noncovalent modification by dendritic molecules. The noncovalent approach is considered more attractive because it is known to preserve the conjugated π-network of CNTs and thus their inherent electronic properties. Noncovalent functionalization of CNTs studied thus far has mainly involved the use of surfactants, aromatic molecules, or linear polymers. The present invention represents the first demonstration of the noncovalent binding of dendrons to CNT sidewalls. Further, the inventors propose fingertip-guided binding as the adsorption mode, where the "fingertips" represent the terminal groups of a dendron that appear to govern the binding efficiency of the molecule to CNT surfaces. The binding mode was revealed by comparing the adsorption efficiency of three types of dendrons that differed only in their terminal groups, Finally, the present inventors demonstrate the spacing ensured by the dendrons and discuss the applicability of the present system to controlled immobilization of biomolecules. The streptavidin-biotin system was used for this investigation.

For the noncovalent functionalization, the present invention selects a dendron having benzyl groups as their termini (Dendron 1 in Fig.1), of which the aromatic benzene rings are expected to bind the sidewalls of CNTs via the π-π interaction. The use of the π-π interaction has been a common approach for the noncovalent functionalization of CNTs. The Dendron 1 was synthesized by amide-coupling of a dendron having carboxylic acid groups with benzylamine. CNTs used in the experiments of the present invention are single-walled CNTs with diameters of 1-2 nm and were grown on SiO₂ZSi wafers by the chemical vapor deposition method as previously reported.⁴² For binding of Dendron 1 onto the sidewalls of the CNTs, CNT-containing wafers were immersed in a 0.20 mM N,N-dimethylformamide (DMF) solution of Dendron 1 for 13 h, Topological images observed by an atomic force microscope (AFM, Nanoscope HIa, Digital Instruments) in tapping mode showed no distinguishable change after the functional ization (Fig. 6), presumably due to the binding nature leading to a minute change in height and the resolution limit of AFM imaging. Hence, a gold nanoparticle is introduced to each immobilized Dendron 2y covalent linkage. After removal of the amine-protecting anthracene group of the dendron molecules attached to the CNT surfaces by treatment with an acid, Tween20 was applied to protect unoccupied (dendron-free) sites of the CNTs from potential nonspecific binding. Tween20 is a surfactant molecule composed of a linear aliphatic chain and three polyethylene oxide (PEO) branches, and is known to bind to CNT surfaces and prevent nonspecific binding. Using N,N'-disuccinimidyl carbonate (DSC) as a coupling reagent, gold nanoparticles having a single amine group (Monoamino Nanogold®, d=2.7 nm, Nanoprobes, inc.) were attached to the dendrons exposing an amine group.

Fig. 2 showed a tapping mode AFM height image, a section profile, and a simplified model of a Dendron 1-treated CNT after applying gold nanoparticles. The AFM section analyses give the height values of the nanoparticles attached to the dendrons consistent with the predicted values. Nanoparticles that appear to be standing upright (point B in Fig. 2) in AFM images have a height value 3.23 ± 0.49 nm higher than the base CNT surface (point A) which is coated only by Tween20. The predicted value is ~3 nm, which is obtained by adding 2.7 nm (the diameter of a nanoparticle) and -1 .3 nm (the estimated height of a dendron when bound to a CNT surface) and subtracting -0.9 ± 0.3 nm (the average height of Tween20 coated on CNTs obtained from independent experiments). To further ensure that the binding of nanoparticles solely involves the covalent bonding between a nanoparticle and a dendron linked by DSC, control experiments were carried out with bare CNTs and Tween20-treated CNTs (Fig, 7). It is certain that no significant binding occurred in either case. It should also be noted that Tween20 does not replace the existing dendrons on the sidewalls of the CNTs, which indicates that the binding of Dendron 1 is more favored than that of Tween20 - a surfactant known to bind to CNTs through the hydrophobic interaction - at least in terms of kinetics.

Next, in order to confirm that the binding of Dendron 1 occurs through the interaction between the terminal groups and the CNT surfaces, comparison experiments were carried out with the Dendron 2 and Dendron 3 which have carboxylic acid and methyl ester groups on their termini, respectively (Fig. 1, (a)). Dendron 2 and Dendron 3 were prepared as previously described, and the same procedure for Dendron 1 was applied for their binding with the wafer-bound CNTs. Fig. 4 shows the tapping mode AFM height images of the samples for each Dendron lfter applying gold nanoparticles. The images clearly show that Dendron 1 and Dendron 2 are successfully bound to the surface while Dendron 3 does not have significant interaction with CNT surfaces. Counting the number of nanoparticles on different nanotubes with various lengths gives the average number per length: 16.8 particles/μm, 24.8 particles/μm, and 5.0 particles/μm for Dendron 1 , Dendron 2, and Dendron 3, respectively. This distinctive result implies that the difference in the adsorption efficiency is attributed to the structure of the dendron, especially to the chemical nature of its terminal groups, namely, the fingertips. The distinction in the binding affinity caused by the different termini also indicates that the single anthracene moiety is less influential on the adsorption event. This is supported by an additional control experiment conducted using a dendron having a t-Boc group instead of the anthracene group of Dendron 1 (Fig.9). The dendron showed a similar binding affinity toward CNTs.

The binding mechanism of Dendron 1 can be explained as a "multiple" π-π interaction. Although previously reported theoretical and experimental studies suggest that the interaction between a single benzene moiety and the sidewall of a CNT is relatively weak, the cooperative action of the multiple benzyl moieties of Dendron 1 is believed to have synergistically enhanced the driving force for the effective binding. The binding of Dendron 2 shows comparable efficiency and this seems to involve the multiple interactions between the carboxylic acids and CNTs. The capability of carboxylic acid groups to hold CNTs through multiple attractions has been studied and utilized for controlled alignment of CNTs on surfaces. Especially in a recent work, the strong attraction was demonstrated both experimentally and theoretically, and depicted as a strong van der Waals interaction. On the contrary, Dendron 3 exhibited weak affinity to the sidewalls of CNTs, which indicates weaker interactions of the methyl ester group with CNTs. The present inventors believe that the weak affinity is due to the relatively weak polarity of the methyl ester group compared to the carboxylic acid group, which is consistent with the previous SAM (self-assembled monolayer) studies, where CNTs favored surfaces having higher polarity.

Next, the present invention investigated the spacing between functionalities provided by the dendrons, by measuring horizontal distances between the highest points of nanoparticles. The smallest spacing was measured to be ~6 nm (Fig. 5) while the average spacing between adjacent nanoparticles (within the distance less than 30 nm) was 14.4 ± 4.9 nm for Dendron 1 and 14.6 ± 5.4 nm for Dendron 2. The present inventors also found that the spacing can be controlled by regulating the number of dendrons immobilized on the nanotube surface. For example, when a higher concentration (1.0 mM) of dendron was used for the same reaction time (13 h), an increased density (34.3 particles/ μm) of the immobilized dendron was observed, from which the smallest spacing was found to be ~3 nm. This value shows a good accordance with the lateral size of the employed dendron.

The same approach was applied to the immobilization of biomolecules. The streptavidin-biotin system was employed for this investigation. Streptavidin has very high affinity for biotin (K₈ ~10¹⁵), and the streptavidin-biotin system has been frequently used for biological and biochemical studies. First, biotin was immobilized at the apex of the dendrons attached to the CNT surfaces, to which streptavidin was allowed to bind. AFM height images reveal that bare CNTs show considerable nonspecific binding, while Tween20-treated CNTs do not (Fig, 8). In contrast, dendron-treated CNTs hold streptavidins in a fairly regulated fashion (average spacing of 14.2 ± 6.1 nm), more importantly with no serious aggregation (Fig. 5, C). The average spacing measured is consistent with the spacing observed for gold nanoparticles, from which the present inventors presume that each streptavidin molecule is linked to a Dendron l nd aggregation was prevented due to the spacing secured by the dendrons.

The average spacing is bigger than that measured for a same-generation dendron that is covalently bound to flat surfaces, in which the strong driving force accompanied by the formation of multiple covalent bonds resulted in compact packing of the dendrons on surfaces. Hence, the covalent approach requires the use of higher-generation dendrons to achieve larger spacing, which involves cumbersome synthetic procedures and generates undesirably large vertical distance of the active functionalities from surfaces. In contrast, the noncovalent approach, which renders relatively small binding energy, generates the larger spacing at the similar concentration and reaction time without relying on a dendron of higher generations. In particular, the average spacing of 14-15 nm has a significant meaning because this value is no smaller than the sizes of most protein molecules commonly found in cells, and at the same time the spacing provides a reasonably high immobilization density on surfaces. Hence, the space ensured by the dendrons would be able to accommodate virtually most kinds of proteins, while avoiding steric hindrance and thus facilitating their specific and selective interactions.

In a biosensor according to the present invention for diagnostic purpose, CNTs are arranged on a substrate, and an electric field of an opposite polarity to a net charge of the receptors is applied to some or all of the CNTs to selectively move receptors for diagnostic target biomolecules to a desired CNTs and to bind them there to a desired position at a high-density.

According to the present invention, suitable materials for the substrate include a variety of polymeric substances, such as silicon, glass, molten silica, plastics, and polydimethylsiloxane (PDMS), and CNTs of several to hundreds of nanometers are arranged on the substrate.

According to the present invention, the receptors are biological substances capable of acting as probes that are detectable when bound to the target biomolecules. Suitable receptors include nucleic acids, proteins, peptides, amino acids, ligands, enzyme substrates, cofactors, and oligosaccharides.

According to the present invention, a target biomolecule, which binds to a receptor, is a biomolecule of interest to be analyzed. The target biomolecule may be proteins, nucleic acids, enzymes, or other boimolecules capable of binding to the receptor.

Assembling streptavidin on the tubes through the dendrons and biotin confirmed the realization of the regulated spacing as well as the elimination of unwanted aggregation. The noncovalent functionalization of CNTs by a Dendron 3an be a new approach toward sensible nanobio-devices, not only by introducing biomolecular probes on CNTs without disruption of the electronic network of the tubes, but also by providing the immobilized probe molecules with an ample space enough to minimize steric hindrance for the unhindered interaction with their target species.

The present invention is further explained in more detail with reference to the following examples. These examples, however, should not be interpreted as limiting the scope of the present invention in any manner.

Example 1: Functionalization of the CNTs

Fig.l (a) Three types of dendrons used for this study, (b) Schematic illustration of the immobilization of gold nanoparticles and streptavidin on CNT surfaces through the fingertip-guided functionalization by dendrons. Chemical identity of the fingertips

(colored red) determines the binding efficiency of a dendron onto CNTs. Dendron 1 and Dendron 2 bind well to CNTs, while Dendron 3 does not. The steps of deprotection and Tween20 treatment are omitted for simplicity.

1.1. Preparation of Dendron Dendrons B and C were prepared as previously described in Hong, B. J.; Oh, S.

J,; Youn, T. 0.; Kwon, S. H.; Park, J. W. Langmitir 2005, 21, 4257-4261).

Dendron 1 was synthesized by amide-coupling of Dendron 2 with benzylamine. Preparation of Dendron 1 (9-anthrylmethylN-({tris[(2-{[tήs({2 [(benzyl amino)carbonyl]ethoxy}methyl)meihylamino]carbonyl}ethoxy)methyl]methylamino}car bonyl)propylcarhamate): 0.50 g (0.31 mmol) of Dendron 2 (9-anthrylmethyl N- ({[tris({2-[({tris[(2- carboxyethoxy)methyl]methyl}amino)carbonyl]ethoxy}methyl)methyl]amino} carbonyOpropylcarbamate, Mw=1614.6) was dissolved in DMF (2.0 ml, Sigma-Aldrich) and a solution of 0.96 g (ca. 15 equiv) of NjN'-dicyclohexylcarbodiimide (DCC, Aldrich), 0.49 mg (ca. 9.5 equiv) of 1-hydroxybenzotriazole (HOBt, Aldrich) in dichloromethane (20 ml, Sigma-Aldrich) was added. After stirring under nitrogen at room temperature for 10 min, 0.32 ml (ca. 9.5 equiv) of benzylamine (Sigma-Aldrich) was added. After stirring for 24 h, dichloromethane and DMF were evaporated. The crude product was dissolved in 100 ml of ethyl acetate (J.T.Baker, HPLC grade) and the solution was washed with 10 % hydrochloric acid and brine. After drying with anhydrous MgSO₄, filtration, and evaporation, the crude product was loaded in a column packed with silica gel. Column chromatographic purification (eluent: ethyl acetateπnethanol = 20:1 (v/v)) resulted in a yellow solid, and the final weight was 0.37 g (Yield = 50 %). ' H NMR(CDCl₃, 300 MHz): δ 8.46 (s, C₁₄H₉CH₂, IH), 8.30 (d, C₁₄H₉CH₂, 2H), 7.99 (d, Ci₄H₉CH₂, 2H), 7.49 (m, C₁₄H₉CH₂, 4H), 7.33-7.17 (m, CH₂C₆H₅, 45Η), 6.67 (s, CH₂CH₂CH₂CONHC, 1Η), 6.49 (t, OCONHCH₂, IH), 6.36 (t, OCH₂CH₂CONHCH₂, 9H), 6.02 (s, Ci₄H₉CH₂O, 2H), 5,85 (s, OCH₂CH₂CONHC, 3Η), 4.34 (d, NHCH₂C₆H₅, 18H), 3.75-3.51 (m, CH₂OCH₂CH₂CONH, 48H), 3.10 (q, NHCH₂CH₂, 2H), 2.32 (t, OCH₂CH₂CONH, 24H), 2.18 (t, CH₂CH₂CH₂CONH, 2H), 1.99-1.88 (m, CH₂CH₂CH₂, 2H). ¹³C NMR (CDCl₃, 300 MHz): δ 172.2 (CH₂CH₂CONH), 157.2 (OCONH), 140.4 (CH₂C₆H₅), 138.1 (Ci₄H₉CH₂), 129.8 (C₁₄H₉CH₂), 128.5 (CH₂C₆H₅), 128.4 (Ci₄H₉CH₂), 127.6 (CH₂C₆H₅), 127.5 (CH₂C₆H₅), 127.3 (Ci₄H₉CH₂), 126.6 (C₁₄H₉CH₂), 126,3 (C₄H₉CH₂), 125.9 (C₁₄H₉CH₂), 69.6 (NHCCH₂O), 69.4 (NHCCH₂O), 67.2 (OCH₂CH₂CONH), 59.7 (C₁₄H₉CH₂), 59.2 (NHCCH₂), 43.5 (CH₂C₆H₅), 40.1 (NHCH₂CH₂), 36.3 (OCH₂CH₂), 33.9 (CH₂CH₂CH₂), 25.7 (CH₂CH₂CH₂). MALDl-TOF MS: m/z = 2438.2 [M+Na]⁺ (exact mass: M = 2415.2).

1.2. Functionalization of Wafer-bound CNTs by dendron CNTs used in thes experiments were single-walled CNTs with diameters of 1 -2 nm and were grown on SiO₂/Si wafers by chemical vapor deposition method as previously reported (Choi, H. C; Kim, W.; Wang, D.; Dai, H. J. Phys. Chem. B 2002, 106, 12361 -12365).

Substrates were immersed in a 0.20 mM N,N-dimethylformamide (DMF) solution of each type of the dendrons for 13 h, followed by rinsing with DMF for 1 min, sonication in DMF, deionized water and methanol each for 30 sec, rinsing with methanol for 1 min, and drying with N₂ stream.

Example 2 :Characterization of the functionalization

Atomic force microscope (AFM) imaging, dispersion experiments, and micro- Raman spectroscopy were employed for the characterization of the functionalization. 2.1 . AFM measurement

All force measurements were performed with a NanoWizard AFM (AFM,

Nanoscope Ilia, Digital Instruments). The spring constant, k_c, of each individual AFM tip was calibrated in solution before each experiment by thermal fluctuation method available via a NanoWizard software. The spring constant varied between 12 and 15 pN/nm, All measurements were carried out in a fresh PBS buffer (pH 7.4) at room temperature. The loading rate of force measurements varied between 1 10 nm/s and 540 nm/s. At each experimental condition, force curves were recorded more than one hundred times at a spot, and at least more than 5 spots were examined. In these measurements, both binding and unbinding force curves were recorded. To calculate distance that the tip actually moved, the cantilever displacement was subtracted from the piezo displacement.

2.2. Dispersion experiments A bath sonicator (Branson 3510) was used for sonication and centrifugation was performed at 2,000 rpm.

The functionalization of CNTs by dendrons was also confirmed by the dispersion experiment. Since the Dendron 1 dissolves in DMF, CNTs functional ized by Dendron 1 is expected to be well dispersed in DMF. HiPco (High-Pressure decomposition of carbon monoxide) single-walled CNTs (0.5 mg) were mixed with Dendron 1 (2.0 mM) in 5.0 ml of DMF and sonicated for 40 min. Then the mixture was centrifuged for 10 min and the supernatant suspension was carefully decanted. A uniform suspension of CNTs was observed in the case of CNTs mixed with Dendron 1 , while the CNTs without the dendron precipitated out during centrifugation ( (a) of Fig. 3). The suspension was stable for ~2 days. This distinct difference in dispersion indicates that the Dendron 1 successfully functional ized the CNTs.

2.3. micro-Raman spectroscopy

Raman spectra were measured for the CNTs bound to Si/SiO₂ wafers with Renishaw Raman system 3000 using Ar ion laser (514.5 nm) as an excitation source,

Raman spectra were analyzed to examine the effect of the functionalization on CNTs. Micro-Raman spectroscopy is one of the most widely used tools to characterize CNTs and their surface modifications. The G-band (~1590 cm^"'), the most intensive high-energy modes of single-walled CNTs, represents the tangential modes which originate from the in-plane stretching modes in graphite. An appearance of the D-band (-1340 cm^'1) in the Raman spectra of CNTs is an indication that the tube surface contains defects, and thus the D-band intensity or the D/G intensity ratio is often used to estimate the degree of covalent functionalization. (b) of Fig.3 shows the Raman spectra obtained before and after the functionalization by the Dendron 1. A very low D-band intensity even after the functionalization process suggests that the functionalization does not involve the disruption of the covalent sp² structures. Fig 3 was (a) Visual comparison of CNTs without (left) and with (right) the Dendron 1 after sonication and centrifugation in DMF. (b) Raman spectra of CNTs before (left) and after (right) the dendron treatment.

Example 3: AFM images of gold NPs attached to dendrons on CNTs

In this example, a gold nanoparticle (NP) was introduced to each immobilized Dendron 2y covalent linkage. After removal of the amine-protecting group of the dendrons by immersing the substrates in 1.0 M dichloromethane solution of trifluoroacetic acid, surfactant Tween20 (1 % in a 50:50 deionized water and DMF mixture) was treated to protect unoccupied (dendron-free) sites of CNTs from potential nonspecific binding. Using N,N'-disuccinimidyl carbonate as a coupling reagent, NPs having a single amine group (Monoamino Nanogold, d=2.7 nm, Nanoprobes) were attached to the dendrons.

Counting the number of NPs on different nanotubes with various lengths gives the average number per length: 16.8 particles/μm, 24.8 particles/μm, and 5.0 particles/μm for Dendron 1 , Dendron 2, and Dendron 3, respectively. The NPs attached to the dendrons were verified by AFM section analyses (Fig. 7). NPs that appear to be standing upright in AFM images have a height value 3.23 ± 0.49 nm higher than the base CNT surface which is coated only by Tween20. The predicted value is ~3 nm, which was obtained by adding 2.7 nm (the diameter of an NP) and ~1 ,3 nm (the estimated height of a dendron when bound to a CNT surface) and subtracting -0,9 ± 0.3 nm (the average height of Tween20 coated on CNTs obtained from independent experiments).

To further ensure that the binding of NPs solely involves the covalent bonding between an NP and a dendron, control experiments were carried out with bare CNTs and Tween20- treated CNTs (Fig. 8). It is certain that no significant binding occurred in either case.

Fig 2 represented (a) An AFM image of a CNT functionalized by Dendron 1 , to which gold nanoparticles are attached, with the height values of three representative parts (see text), (b) Section profiles and (c) the corresponding model of the parts designated in (a). Height values in white color (and without brackets) in (a) indicate the heights from the bottom of the substrate, and the values in orange color (and with brackets) are the differences from the point A.

3.1. Deprotection of the immobilized dentrons

The substrates were immersed into a dichloromethane solution of 1.0 M trifluoroacetic acid (TFA). After 3 h, they were transferred into a dichloromethane solution of 20 % (v/v) diisopropylethylamine (DIPEA) and left for 10 min. The substrates were then sonicated in dichloromethane and methanol each for 30 sec, in fresh methanol again for 30 sec, rinsed with methanol for 1 min, and dried with N₂ stream.

3.2. Treatment with Tween 20

The substrates were soaked in a 1 % (v/v) solution of Tween20 in a 50:50 mixture of deionized water and DMF for 3 h. The substrates were then sonicated in deionized water, dichloromethane, and methanol each for 30 sec, rinsed with methanol for 1 min, and dried with N₂ stream. The result was shown in Fig. 7 3.3. N,N'-disuccinimidyl carbonate linker reaction

The substrates were placed in an acetonitrile solution of DSC (25 mM) and diisopropylethylamine (DlPEA, 1 mM) for 4 h under N2 atmosphere. The substrates were then washed with DMF for 1 min and methanol for 30 sec, and dried with N₂ stream.

3.4. Applying gold nanoparticle

A NaHCO₃ (20 1, 50 mM, pH 8.5, 10 % dimethyl surfoxide (DMSO)) solution of gold nanoparticles (6 μM) was dropped onto each substrate kept in a saturated humidity chamber, and 8 h was allowed for the reaction. The substrates were washed with deionized water and DMSO each for 1 min, then placed in stirred DMSO for 2 h, rinsed with DMSO and methanol each for 1 min, and dried with N₂ stream.

3.5. Counting the number and the height of gold nanoparticle In a typical analysis, 3-4 different spots of 5μm x 5μm area were scanned with

AFM for a sample, and several tubes that appear in the area were investigated. Images of l μm ^χ l μm or often 500 nni x 500 nm along each tube were subsequently obtained and used for counting the number. The analysis was carried out for multiple samples from independent batches. The numbers of particles used for the spacing analyses were 68 and 84 for Dendron 1 and Dendron 2, respectively. Heights of nine vertically standing particles were taken to obtain the average height value. Because measuring the height of particles at the specific orientation (such as point B in Fig. 2) is desirable, a limited number of particles were taken into the calculation.

Fig. 2 shows (a) an AFM image of a CNT functional ized by Dendron 1 , to which gold nanoparticles are attached, with the height values of three representative parts (see text), (b) Section profiles and (c) the corresponding model of the parts designated in (a). Height values without brackets in (a) indicate the heights from the bottom of the substrate, and the values with brackets are the differences from the point A.

Fig. 4 showed AFM images of CNTs functionalized by dendrons, to which gold nanoparticles are applied, showing discriminative binding efficiency of Dendron 1 , Dendron 2 and Dendron 3. CNTs were treated with (a) Dendron 1, (b) Dendron 2, and (c) Dendron 3, respectively.

Fig. 5 was spacing realized by the dendrons. (a) An example of small spacing between nanoparticles (~6 nm) for Dendron 1. The distance was measured to be the horizontal distance between the highest points of two NPs, indicated by arrows, (b)

Distribution of the spacing between adjacent nanoparticles for Dendron 1 (top) and Dendron 2 (bottom), (c) Streptavidins on a CNT modified by Dendron 1.

Example 4: Immobilization of streptavidin

4.1. Immobilization of streptavidin

After the same steps for the gold nanoparticle attachment were followed up to the DSC linker reaction, 20 mL Of NaHCO₃ solution (50 mM, pH 8.0) dissolving (+)- biotinyl-3,6-dioxaoctanediamine (EZ-Link® Biotin-PEO-Amine, Pierce; 0.5 mg/ml) was placed on top of a substrate, and subsequently 3 h at room temperature was allowed for the reaction between the NHS group and the amine group of the biotin. After washing with deionized water briefly and drying with a stream of nitrogen, the treated substrate was placed in 20 mL of NaHCO₃ solution (50 mM, pH 8.0) dissolving ethanolamine (0.5 % v/v). Three hours were allowed at room temperature to quench the remained NHS group. Subsequently, brief washing of the resulting substrate with DMF was followed. Finally, a phosphate buffer saline solution (20 mL, 10 mM, pH 7.4) dissolving streptavidin (Sigma-Aldrich; 0.010 mg/ml) was placed on top of the substrate, and 30 min was allowed at room temperature. The substrate was placed in deionized water with stirring for 1 h, and dried with N₂ stream. 4.2. Control experiments with streptabivin

Control experiments with streptavidin on (a) bare CNTs and (b) Tween20-coated CNTs were carried out. In Fig. 8, (a) The image shows considerable nonspecific binding of streptavidin onto CNTs. (b) The image reveals that Tween20 repels streptavidin.

Example 5: A control test with t-Boc protected Dendron 1 analog

A Dendron 3ontaining a t-Boc moiety as the protecting group instead of the anthryl group was tested to investigate the influence of the latter protecting group on the binding event. While the anthryl group might interact with CNTs through π-π interaction, the t-Boc group is not the case. A dendron having the same terminal groups as those of Dendron 1 and a t-Boc group as the protecting group (Fig. 9, a) was employed for this investigation, and showed a similar binding probability or density (18.1 particles/μm).

The procedures for the immobilization of Dendron 1 ' on CNTs were the same as those for Dendron 1. NMR and mass spectral data of Dendron 1 '

¹H NMR(CDCl₃, 300 MHz) δ 7.28-7.23 (m, CH₂C₆H₅, 45H), 7.21-7.18 (s, NHCH₂C₆H₅, 9H), 6.83 (s, CH2CH2CH2CONHC, 1Η), 6.64 (s, OCΗ2CΗ2CONHC, 3Η), 6.08 (s, (CH₃)₃OCONH, 1 Η), 4.34 (d, NHCH₂C₆H₅, 18H), 3.65-3.52 (m, CH₂OCH₂CH₂CONH, 48H), 3.01 (q, NHCH₂CH₂, 2H), 2.40-2.35 (t, OCH₂CH₂CONH, 18H), 2.20-2.18 (t, OCH₂CH₂CONH, 6H), 2.1 1 (t, CH2CH2CH2CONH, 2H), 1.63 (m, CH2CH2CH2, 2H), 1.41 (s, (CH₃)₃, 9Η).

¹³C NMR (CDCl₃, 300 MHz) δ 171.4 (CH₂CH₂CONH), 171.2 (OCONH), 138.5 (CH₂C₆H₅), 128.6-127.3 (CH₂C₆H₅), 69.4 (NHCCH20), 69.2 (NHCCH2O), 67.3 (OCH2CH2CONH), 59.2 (NHCCH2), 43.4 (CH₂C₆H₅), 36.4 (OCH2CH2), 28.4 ((CH₃)₃).

MALDI-TOF MS: m/z = 2304.2 [MH-Na]⁺ (exact mass: M = 2281.2).

Claims

WHAT IS CLAIMED IS:

1. A carbon nanotube (CNT)-Dendron composite where a plurality of termini of a branched region of at least a Dendron comprising a branched region and a linear region are bound non-covalently to the sidewall of CNT.

2. The CNT-Dendron composite of claim 1, wherein the dendron is a chemical compound having the following chemical structures.

Chemical formula 1

Where, Rl is OH₁ or _N

R2 is a protecting group of tert-butyloxycarbonyl or

3. The CNT-Dendron composite of claim 1, wherein the dendrons are spaced at regular intervals between about 0,1 nm and about 100 nm among the linear regions.

4. The CNT-Dendron composite of claim 3, wherein the dendrons are spaced at regular intervals between about 10 nm and about 30 nm among the linear regions.

5. The CNT-Dendron composite of claim 1, wherein the CNT of the composite maintains its π-configuration.

6. The CNT-Dendron composite of claim 1, wherein the CNT is multi-wall nanotube structures (MWNT) or single-wall nanotube structures (SWNT).

7. The CNT-Dendron composite of claim 6, wherein the CNT is completely, partially or not at all oxidized.

8. A biosensor for detecting a biomolecule comprising a carbon nanotube (CNT)- Dendron composite which a plurality of termini of a branched region of at least a Dendron comprising a branched region and a linear region are bound non-covalently to the sidewall of CNT.

9. The biosensor for detecting a biomolecule of claim 8, wherein the dendrons provide a binding site for a receptor for a target biomolecule.

10. The biosensor for detecting a biomolecule of claim 8, wherein the substrate is formed of a material selected from the group consisting of silicon, glass, molten silica, plastics, and polydimethylsiloxane (PDMS).

1 1 . The biosensor for detecting a biomolecule of claim 9, wherein the receptor is selected from the group consisting of nucleic acids, proteins, peptides, amino acids, ligands, enzyme substrates, cofactors, and oligosaccharides.

12, The biosensor for detecting a biomolecule of claim 9, wherein the target biomolecule is selected from the group consisting of proteins, nucleic acids and enzymes.