Controlled Synthesis of Double- and Multiwall Silver Nanotubes with

When the EDAG gel was mixed with AgNO3, the vibrations ascribed to the hydrogen bond between carboxylic acid such as those at 1729, 1695, 1678, and 15...
0 downloads 0 Views 326KB Size
Download PDF
Langmuir 2006, 22, 775-779

775

Controlled Synthesis of Double- and Multiwall Silver Nanotubes with Template Organogel from a Bolaamphiphile Peng Gao, Chuanlang Zhan, and Minghua Liu* CAS Key Laboratory of Colloid and Interface Science, Institute of Chemistry, The Chinese Academy of Sciences, Beijing 100080, People’s Republic of China ReceiVed July 1, 2005. In Final Form: NoVember 16, 2005 Double- and multiwall silver nanotubes were synthesized by using the uniform low-molecular-mass organogel nanotubes self-assembled from an L-glutamic-acid-based bolaamphiphile, N,N-eicosanedioyl-di-L-glutamic acid (EDGA). The EDGA could gel a mixed water/ethanol solvent and form helical nanotubes. When the gel thus formed was mixed with AgNO3 in water/ethanol, the silver(I) cations could be coordinated with both the inner and outer surfaces of the EDGA nanotubes. The reduction of the silver cation under the photoirradiation yielded double-wall silver nanotubes, where two silver layers were separated by one EDGA layer. Elongations of the reduction time of the mixed gels and AgNO3 in the solution lead to the formation of three-, four-, and five-wall silver nanotubes. In these multiwall silver nanotubes, each wall was separated at a distance of about 2.7 nm, which was just the molecular length of the bolaamphiphile. It was suggested that the dissolved EDGA molecules and excess AgI cations were further assembled onto the surface of the formed double-wall silver nanotubes and, as a consequence, the photoreduction caused the formation of the third-wall silver nanotubes. The multiwall silver nanotubes were further formed in a similar way. The factors affecting the formation of the silver wall nanotubes such as the relative amount of AgNO3 to EDGA and the synthetic conditions were discussed.

Introduction Nanotubes such as carbon nanotubes,1 gold nanotubes,2 silver nanotubes,3 nanotubes of other metals,4 and polymers5 have attracted increasing interests because of their remarkable physiochemical properties and potential applications in catalysis, separations of biomolecules, and fabricating nanoscale electronic, optoelectronic, and magnetic devices.1-5 Among various methods for preparing the nanotubes with different compositions and morphologies, the template method, which entails the synthesis of a desired material in a controlled process, has been proven to be a versatile and inexpensive technique.6 Diverse templates such as nanoporous alumina,7 polycarbonate filters,8 selfassembled lipid nanotubes,9,10 and orgnaogels11 have been used toward the synthesis of the nanotubes. * To whom correspondence should be addressed. Telephone: +86-1082612655. Fax: +86-10-62569564. E-mail: [email protected]. (1) (a) Iijima, S. Nature 1991, 354, 56. (b) Che, G.; Fisher, E. R.; Martin, C. R. Nature 1998, 393, 346. (2) (a) Yu, S.; Lee, S. B.; Martin, C. R. Anal. Chem. 2003, 75, 1239. (b) Lee, S. B.; Martin, C. R. J. Am. Chem. Soc. 2002, 124, 11850. (c) Jirage, K. B.; Hulteen, J. C.; Martin, C. R. Anal. Chem. 1999, 71, 4913. (d) Jirage, K. B.; Hulteen, J. C.; Martin, C. R. Science 1997, 278, 655. (e) Mishizawa, M.; Menon, V. P.; Martin, C. R. Science 1997, 268, 700. (3) (a) Zhang, S. H.; Xie, Z. X.; Jiang, Z. Y.; Xu, X.; Xiang, J.; Huang, R. B.; Zheng, L. S. Chem. Commun. 2004, 1106. (b) Kijima, T.; Yoshimura, T.; Uota, M.; Ikeda, T.; Fujikawa, D.; Mouri, S.; Uoyama, S. Angew. Chem. Int. Ed. 2003, 43, 228. (4) (a) Hu, P.; Liu, Y.; Fu, L.; Cao, L.; Zhu, D. Chem. Commun. 2004, 556. (b) Bao, J.; Wang, K.; Xu, Z.; Zhang, H.; Lu, Z. Chem. Commun. 2003, 208. (c) Wang, J.; Li, Y. AdV. Mater. 2003, 15, 445. (d) Cao, M.; Hu, C.; Wang, Y.; Guo, Y.; Guo, C.; Wang, E. Chem. Commun. 2003, 1884. (e) Crowley, T. A.; Ziegler, K. J.; Lyons, D. M.; Erts, D.; Olin, H.; Morris, M. A.; Holmes, J. D. Chem. Mater. 2003, 15, 3518. (5) (a) Demoustier-Champagne, S.; Stavaux, P. Y. Chem. Mater. 1999, 11, 829. (b) Sukeerthi, S.; Contractor, A. Q. Anal. Chem. 1999, 71, 2231. (6) (a) Martin, C. R. Science 1994, 266, 1961. (b) Hulteen, J. C.; Martin, C. R. J. Mater. Chem. 1997, 7, 1075. (c) Martin, C. R.; Mitchell, D. T. Anal. Chem. 1998, 70, 322. (d) Wirtz, M.; Parker, M.; Kobayashi, Y.; Martin, C. R. Chem.s Eur. J. 2002, 8, 3573. (7) Hornyak, G. L.; Patrissi, C. J.; Martin, C. R. J. Phys. Chem. B 1997, 101, 1548. (8) Nishizawa, M.; Menon, V. P.; Martin, C. R. Science 1995, 268, 700. (9) (a) Spector, M. S.; Price, R. R.; Schnur, J. M. AdV. Mater. 1999, 11, 337. (b) Schnur, J. M.; Price, R.; Schoen, P., et al. Thin Solid Films, 1987, 152, 181.

Low-molecular-mass organogelators (LMOGs) are an important class of soft matter, which could form exquisite 1D nanostructures and then imprison a large quantity of liquid.11 The 1D nanostructures formed through the gelation generally include the nanofibers, nanoribbons, and nanotubes. These organized nanostructures could be easily used as templates or structure-directing agents for the synthesis of inorganic, semiconductor, and metal nanostructures.12-18 The organogelators usually contain hydrophilic groups, and many metal cations can be adsorbed or coordinated onto the 1D nanostructures formed from the organogelators. Subsequent reduction of the metal cations or reaction with appropriate anions brought out the formation of nanostructured metals or semiconductors. For example, CdS semiconductors were synthesized by the mineralization of CdII (10) Price, R. R.; Dressick, W. J.; Singh, A. J. Am. Chem. Soc. 2003, 125, 11259. (11) (a) Terech, P.; Weiss, R. G. Chem. ReV. 1997, 97, 3133. (b) Gronwald, O.; Shinkai, S. Chem.sEur. J. 2001, 7, 4329. (c) van Bommel, K. J. C.; Friggeri, A.; Shinkai, S. Angew. Chem. Int. Ed. 2003, 42, 9. (12) (a) Jung, J. H.; Ono, Y.; Hanabusa, K.; Shinkai, S. J. Am. Chem. Soc. 2000, 122, 5008. (b) Ono, Y.; Kanekiyo, Y.; Inoue, K.; Hojo, J.; Shinkai, K. Chem. Lett. 1999, 475. (c) Ono, Y.; Nakashima, K.; Sano, M.; Hojo, J.; Shinkai, S. Chem. Lett. 1999, 1119. (d) Sugiyasu, K.; Tamaru, S.; Takeuchi, M.; Berthier, D.; Huc, Ivan.; Oda, R.; Shinkai, S. Chem. Commun. 2002, 1212. (e) Jung, J. H.; Kobayashi, H.; van Bommel, K. J. C.; Shinkai, S.; Shimizu, T. Chem. Mater. 2002, 14, 1445. (f) Jung, J. H.; Ono, Y.; Shinkai, S. Langmuir 2000, 16, 1643. (g) Jung, J. H.; Ono, Y.; Shinkai, S. Chem.sEur. J. 2000, 6, 4552. (h) Jung, J. H.; Kobayashi, H.; Masuda, M.; Shimizu, T.; Shinkai, S. J. Am. Chem. Soc. 2001, 123, 8785. (i) Jung, J. H.; Nakashima, K.; Shinkai, S. Nano Lett. 2001, 1, 145. (j) Jung, J. H.; Lee, S. H.; Yoo, J. S.; Yoshida, K.; Shimizu, T.; Shinkai, S. Chem.sEur. J. 2003, 9, 5307. (13) Gundiah, G.; Mukhopadhyay, S.; Tumkurkar, U. G.; Govindaraj, A.; Maitra, U.; Rao, C. N. R. J. Mater. Chem. 2003, 13, 2118. (14) Yang, Y. G.; Suzuki, M. Shirai, H.; Kurose, A.; Hanabusa, K. Chem. Commun. 2005, 2032. (15) (a) Xue, P.; Lu, R.; Huang, Y.; Jin, M.; Tan, C.; Bao, C.; Wang, Z.; Zhao, Y. Langmuir 2004, 20, 6470. (b) Xue, P.; Lu, R.; Li, D.; Jin, M.; Tan, C.; Bao, C.; Wang, Z.; Zhao, Y. Langmuir 2004, 20, 11234. (16) Sone, E. D.; Zubarev, E. R.; Stupp, S. I. Angew. Chem. Int. Ed. 2002, 41, 1705. (17) Llusar, M.; Pidol, L.; Rous, C.; Pozzo, J. L.; Sanchez, C. Chem. Mater. 2002, 14, 5124. (18) Zhan, C.; Wang. J.; Yuan, J.; Gong, H.; Liu, Y.; Liu, M. Langmuir 2003, 19, 9440.

10.1021/la0517787 CCC: $33.50 © 2006 American Chemical Society Published on Web 12/20/2005

776 Langmuir, Vol. 22, No. 2, 2006 Scheme 1. Molecular Structure of the Bolaamphiphile Used in This Work

Gao et al.

the transmission electron microscopy (TEM) as well as electron diffraction (ED). FT-IR spectra were also used to characterize the nanotubes. Experimental Procedures

cations on supramolecular ribbons.16 We have used nanofibers self-assembled from a racemic gelator, 2-acryloylamide-dodecane-1-sulfonic acid, as a template and synthesized both rightand left-handed silver nanohelices.18 In this paper, we employed the organogels formed from a bolaamphiphile and synthesized double- and multiwall silver nanotubes (DWSNTs and MWSNTs, respectively) in a controlled way. Bolaamphiphiles describe those molecules in which two polar functional headgroups are linked covalently by one or more hydrophobic hydrocarbon chains.19-23 In comparison with the conventional amphiphile (single chain, single headgroup), bolaamphiphiles have many interesting propereties.15-19 For example, a bolaamphiphile could possibly take three kinds of conformations at the air/water interface and afford the possibility to form various surface nanoarchitectures.21 Some of the bolaamphiphiles were found to form excellent organogels with nanofiber or nanotube structures. Many excellent examples are summarized in two recent comprehensive reviews.22a,24 One important feature of the bolaamphiphile is that it has two hydrophilic headgroups, which promote the formation of the stable monolayer lipid membranes (MLMs) in the organized nanostructure. Recently, we have observed that an L-glutamicacid-based bolaamphiphilic LMOG, N,N-eicosanedioyl-di-Lglutamic acid (EDGA, Scheme 1), can gel in a 1:1 mixture of ethanol/water and self-assemble into uniform helical nanotubes with a tubular wall thickness of ∼3 nm, which is equal to the full length of EDGA.25 The helical structure was suggested to be rolled from a stable MLM. Therefore, both of the surfaces of the EDGA nanotubes are hydrophilic and are expected to show strong affinity for metal cations. In this paper, we used such helical nanotube as a template and investigated the synthesis of silver nanotubes. It was found that the EDGA nanotubes showed enough stability in hydrophilic solvents such as acetone, ethanol, and a 1:1 mixture of ethanol/water. When the preformed gel was mixed with AgNO3 in 1:1 ethanol/water, the AgI cation can be coordinated with the surface of the gel, which served as a ligand. Upon photoirradiation, the AgI cation could be reduced to a metal silver. DWSNTs with clear two layers of silver and one layer of amphiphile were thus obtained. Furthermore, we have also obtained three-, four-, and five-wall silver nanotubes with longer time photoirradiation. In these MWSNTs, each silver wall was separated from another by just a single EDGA layer. A possible mechanism was proposed to explain the formation of such DWSNTs and MWSNTs. The formation process of the silver nanotubes was monitored by the UV-vis spectral measurements, and the silver nanotubes were characterized by (19) Escamilla, G. H.; Newkome, G. R. Angew. Chem. Int. Ed. 1994, 33, 1937. (20) Masuda, M.; Shimizu, T. Langmuir 2004, 20, 5969. (21) (a) Menger, F. M.; Keiper, J. S. Angew. Chem. Int. Ed. 2000, 39, 1906. (b) Kohler, K.; Forster, G.; Hauser, A.; Dobner, B.; Heiser, U. F.; Ziethe, F.; Richter, W.; Steiniger, F.; Drechsler, M.; Stettin, H.; Blume, A. J. Am. Chem. Soc. 2004, 126, 16804. (c) Menger, F. M.; Littau C. A. J. Am. Chem. Soc. 1991, 113, 1451. (d) Iwaura, R.; Yoshida, K.; Masuda, M.; Yase, K.; Shimizu, T. Chem. Mater. 2002, 14, 3047. (22) (a) Fuhrhop, J.-H.; Wang, T. Chem. ReV. 2004, 104, 2901. (b) Fuhrhop, J.-H.; Fritsch, D. Acc. Chem. Soc. 1986, 19, 130. (c) Fuhrhop, J. H.; Matthieu, J. Angew. Chem. Int. Ed., 1984, 23, 100. (23) Fyles, T. M.; Zeng, B. J. Org. Chem. 1998, 63, 8337. (24) Shimizu, T.; Masuda, M.; Minamikawa, H. Chem. ReV. 2005, 105, 1401. (25) Zhan, C.; Gao, P.; Liu, M. Chem. Commun. 2005, 462.

Materials. The bolaamphiphile, EDGA, was synthesized by the amidation of L-glutamic acid diethyl ester with eicosanedioic acid and followed by the hydrolysis of tetraethyl N,N-eicosanedioyl-diL-glutamic esters to give the free acid.25 All of the other chemical reagents were analytical-grade and used without further purification. Millipore-Q water (18 MΩ cm) was used in all of the cases. Gel Formation and Its Stability. A total of 1 mL of EtOH/H2O (1:1 in a volume ratio) containing 3.5 mg of EDGA was put into a test tube and heated at 75 °C for 30 min to give a transparent solution. Then, the solution was removed and left to cool at room temperature. A homogeneous organogel was obtained after 2 h. A total of 1/10 of the formed organogel was dispersed into 5 mL of various solvents such as the mixed solvent of EtOH/H2O (1:1 molar ratio), acetone, THF, and so on, at 20 °C to check the stability of the gel in the mixed solvent. It was found that the gel nanostructures could be kept in the mixed ethanol/water for more than 16 h under dark. Synthesis of the Silver Wall Nanotubes. The EDGA gel was first dispersed into the mixed ethanol/water solvent for 4 h, and then various amounts of AgNO3 were added to the gel dispersions. The mixture was degassed with pure nitrogen for 15 min and further stirred under dark for 12 h. After that, the dispersion was allowed to UV-irradiate (λ ) 254 nm, 25W) for 1 h, followed by the irradiation of daylight. In different time intervals of the dispersion under daylight, a certain part of the dispersion was taken out and subjected to the centrifugation (3600 rpm for 5 min). The sediment was washed with 5 mL of EtOH, stirring softly for 5 min and then put to centrifugation and washed again. Thus, resulted silver nanotubes were dried under vacuum at room temperature for 30 min. The dried sample was then dispersed into a mixture of water and ethanol (in a 1:1 volume ratio) in darkness, and a drop of the dispersion was cast on the copper grid for the TEM measurement. To obtain good silver nanotubes, the mixed dispersion of the gel and AgNO3 was kept unstirred after the irradiation was initiated. Procedure. A piece of freshly cleaved mica was dipped into the EDGA solution and allowed to form organogels. This mica was put into vacuum to remove the solvent for the AFM measurement. The dispersed gel in ethanol/water and its mixture with AgNO3 after a certain time were cast onto freshly cleaved mica or CaF2 for AFM and FT-IR measurements, respectively. All of these cast films were put to vacuum to remove the solvent before measurement. UV-vis spectra were used to monitor the photoreduction of the gel/AgNO3 dispersions in ethanol/water. Instruments. AFM was performed using Tapping Mode (Nanoscope IIIa, Digital Instruments, Inc.) with a pyramidal Si3N4 tip. TEM was observed using a JEOL 300 system, which operated at 200 kV. FT-IR and UV-vis spectra were measured with a JASCO FT/IR-660 plus spectrometer and a JASCO 530 UV-vis spectrometer, respectively.

Results and Discussion 1. Stability of Helical Nanotubes in Solutions. EDGA could gel in mixed solvents of water and ethanol and form the helical nanotubes with a wall thickness of about 3 nm, as shown in Figure 1a.25 However, if we mixed EDGA and AgNO3 in solution, no gel could be formed except the precipitation. To use the nanotube of the organogel as the template to synthesize the silver nanotubes, we mixed the preformed EDGA with AgNO3. Before mixing with AgNO3, we dispersed the EDGA gel into various solvents and checked their stability first. Figure 1b shows the AFM picture of the cast film from the EDGA gel dispersion in the ethanol/water after 16 h. It is clear that the helical structure of the EDGA gel still kept its basic morphology after dispersing in the solvent for 16 h, although some dissolved molecular

Synthesis of Double- and Multiwall SilVer Nanotubes

Langmuir, Vol. 22, No. 2, 2006 777

Figure 1. (a) AFM images of the as-formed EDGA organogel with 1:1 mixed water/ethanol. (b) Organogels after dispersed into a 1:1 mixture of water/ethanol for 16 h. (c) EDGA gels after mixed with AgNO3 under dark for 4 h.

Figure 2. UV-vis spectra of the mixed solution of EDGA and AgNO3 at different exposed time to the photoirradiation.

aggregates could be detected. Besides the ethanol/water-mixed solvent, we have also tested other solvents such as THF, CHCl3, or acetone. The helical nanostructures of the gel could be easily destroyed in these solvents, or the gel could not be dispersed. Therefore, we selected the 1:1 mixture of water/ethanol as a dispersion solvent. In addition, the fiber structure could be kept even mixed with AgNO3, although the gels were bundled to become larger, as shown in Figure 1c. 2. Formation of the DWSNTs. UV-Vis Spectra. The formation of the silver nanotube under the presence of the EDGA nanotube template was monitored by the UV-vis spectra. Figure 2 shows the UV-vis spectral changes of the EDGA and AgNO3 solution upon exposure to the UV irradiation of 254 nm and daylight. After 20 min of exposure to UV light, weak plasma absorption with a peak maximum at 401 nm was observed, indicating the formation of silver nanoparticles. This peak increased with the exposure time. Meanwhile, another peak at 547 nm increased as well. After 6 h, these two peaks at 401 and 547 nm were clearly observed, suggesting that the silver nanotubes might have been formed. These two peaks could be attributed to the transverse and the longitudinal modes of the silver 1D nanostructure, respectively.26 The broad absorption band in the region from 350 to 800 nm is considered to stem from the coupling of electromagnetic waves between neighboring nanotubes. TEM Characterization. The mixed solution of EDGA gel and AgNO3 subjected to photoirradiation at different time intervals was taken out, and the formed silver nanotubes were purified. The nanotubes formed in different conditions were observed with TEM. Parts a-c of Figure 3 show the TEM images of the formed nanotubes after exposing the mixtures of the EDGA gel and AgNO3 (EDGA/AgNO3 in a 3.5:5 weight ratio) for 17 h. Many nanotubes nearly uniform in size are observed (Figure 3a). The nanotube has a diameter of about 50 nm, and some of them can be extended to several micrometers. In the enlarged TEM image as shown in parts b and c of Figure 3, three layers were clearly observed for the nanotubes. The central bright part can be suggested as the hollow sphere of the nanotubes, while the (26) (a) Bockrath, M.; Liang, W.; Bozovic, D.; Hafner, J. H.; Lieber, C. M.; Tinkha, M.; Park, H. Science 2001, 291, 283. (b) Murphy, C. J. Science 2002, 298, 2139. (c) Dickson, R. M.; Lyon, L. A. J. Phys. Chem. B 2000, 104, 6095.

Figure 3. TEM images of the SNTs synthesized from the mixed solution (EDGA/AgNO3 in a 3.5:5 weight ratio) at the exposed time to daylight for 17 h without stirring (a-c) and for 30 h (e and f). TEM image of the as-formed EDGA organogel with 1:1 mixed water/ethanol (d). The inset in b is the ED pattern taken from the tubular wall of the DWSNT. The inset in d is the ED pattern taken from the EDGA organogel nanotube.

dark layers could be regarded as the silver wall. To confirm this, we have examined the electronic diffraction (ED) of the dark places, as shown in the inset of Figure 3b. In comparison, we have also taken the ED pattern of the pure EDGA nanotube. Hexagonal diffraction spots are clearly observed in the nanotubes synthesized by the reduction of AgNO3 in the EDGA gels, while no pattern was observed for EDGA gel alone. This confirmed that silver wall nanotubes were indeed formed using the photoreduction of AgNO3 in gel dispersion. Furthermore, it is obvious from the ED pattern that the formed silver wall nanotube has a monocrystallinity. It should be noted that because of the higher surface energy of the as-reduced silver, it will aggregate the surrounding AgI cations, which were further reduced to the silver clusters on the silver wall. Therefore, in the HRTEM, the silver wall seemed to be composed of nanoparticles. Furthermore, careful observation on the silver nanotube reveals that the wall consists of two layers. Each of the layers was separated by a distance of ca. 2.7 nm. Therefore, we can name this silver nanotube as a DWSNT. Considering the fact that the EDGA molecule has a molecular length of ca. 3.0 nm, it can be suggested that the two silver walls are separated by just one EDGA molecules, in which the alkyl chain is slightly titled. The silver cation has a strong affinity for the carboxylic acid group. Because both the inner and outer surfaces of the EDGA nanotubes are composed of the carboxylic acid groups, it can be assumed that the silver(I) cations were adsorbed or coordinated with the

778 Langmuir, Vol. 22, No. 2, 2006

Gao et al.

Figure 4. (a) TEM image of the synthesized silver nanoparticles (EDGA/AgNO3 in a 3.5:5 weight ratio) at the exposed time to daylight for 17 h with stirring. (b) TEM image of the synthesized SNTs (EDGA/AgNO3 in a 0.35:1 weight ratio) at the exposed time to daylight for 17 h without stirring.

EDGA first and reduced to the DWSNT subsequently upon photoirradiation. When the mixed solution of AgNO3 and the EDGA organogel was irradiated for a longer time, nanotubes with a multilayered wall were obtained. Parts e and f of Figure 3 show the TEM pictures of the nanotubes obtained after 30 h of irradiation. As shown in Figure 3e, the formed nanotube has a thickness of about 50 nm, while the wall was composed of three layers. It was further observed that each layer was separated by a distance of 2.7 nm, indicating that the silver walls were separated by one EDGA layer. Besides the three-wall silver nanotubes, we have also observed a few four- and five-layer silver wall nanotubes, as shown in Figure 3f. However, in these MWSNTs, the wall was not as perfect as DWSNTs. It should be noted that to get the well-defined DWSNTs and MWSNTs it is important to keep the solution undisturbed during the photoirradiation. If we continued to stir the dispersion during the photoirradiation, only nanoparticles could be obtained, as shown in Figure 4a. This indicated that stirring would destroy the ordered template gel. Although the EDGA gel itself was stable in the mixed solvent of ethanol and water, the adsorption and/or bonding of the AgI cation on the gel would accelerate the destruction of the template organogel and only lead to the formation of nanoparticles. On the other hand, the relative amount of EDGA and AgNO3 is also an important factor for the formation of the silver wall nanotubes. In our experiments, we mainly used the dispersions containing 0.35 mg of EDGA and 0.5 mg in a 5-mL solvent, where the relative molar ratio of EDGA/AgI is 1:5. When less AgNO3, for example, 2.5 M AgNO3 with respect to 1 M EDGA, was used, no DWSNTs were obtained. When excess AgNO3, for example, 10 M AgNO3 with respect to 1 M EDGA, was used, no perfect DWSNTs were obtained. It was observed that excess AgNO3 was reduced into silver nanoparticles, which aligned along the wall of the DWSNTs, as shown in Figure 4b. 3. Spectral Confirmation. To further investigate the formation mechanism of the silver nanotubes, we have compared the FTIR spectra of the EDGA with those of the silver nanotubes. Figure 5 shows the FT-IR spectrum of the EDGA gel and those mixed with AgNO3 and the formed DWSNTs. In the FT-IR spectrum of the gel from the water/ethanol solvent, strong vibration bands were observed at 3327, 2918, and 2849 cm-1 and a series of vibrations in the region of 1800-1400 cm-1. The band that appeared at 3327 cm-1 can be ascribed to the amide stretching of the EDGA molecules, in which the amide group was suggested to form a hydrogen bond. Strong vibration bands that appeared at 2918 and 2849 cm-1 can be assigned to the symmetric and antisymmetric stretching vibration of CH2, which indicated that the alkyl chains packed in a zigzag all-trans conformation in the gel. Strong vibrations observed at 1729, 1695, 1678, and 1594 cm-1 can be assigned to carbonyl stretching

Figure 5. FT-IR spectra of EDGA gel, its mixture with AgNO3 and EDGA, and the formed silver nanotubes.

with the hydrogen bond between carboxylic acid, while the band at 1641 and 1541 cm-1 can be assigned to the CdO and NH stretching of the amide groups, respectively, which are hydrogenbonded. When the EDAG gel was mixed with AgNO3, the vibrations ascribed to the hydrogen bond between carboxylic acid such as those at 1729, 1695, 1678, and 1594 cm-1 disappeared, while the bands at 1641 cm-1 showed a slight shift to 1645 cm-1. The band at 1541 cm-1 in EDGA broadened after mixing with AgNO3, and a strong band appeared at 1529 cm-1. This indicated that, upon mixing with AgI, the hydrogen bond between carboxylic acid was destroyed, while the hydrogen bond between the amide groups was kept. The destruction of the carboxylic acid was further verified from the appearance of a new strong band at 1529 cm-1 and a band at 1402 cm-1. These two bands can be assigned to the antisymmetric and symmetric stretching vibrations of carboxylate, respectively. It is well-known that in the FT-IR spectrum the separation (∆) between the antisymmetric and symmetric stretching vibrations of carboxylate [∆ ) νas(COO) - νs(COO)] could be used as a diagnosis to determine the coordination type (ionic, monodentate, bridging bidentate, or chelating bidentate) between the carboxylate and metal ions.27,28 Although there are slight differences among various literatures, the relationship between the separation values and the coordination types can be generalized as follows: ionic (164 cm-1), monodentate (200-300 cm-1), bridging bidentate (140-170 cm-1), and chelating bidentate (40-110 cm-1).27,28 In the present case, the separation was 127 cm-1, suggesting that AgI was coordinated with the carboxylate. These results indicated that, when the EDGA gel was mixed with AgNO3, the carboxylic acid reacted with silver(I) ion. However, because of the hydrogen bond between the amide groups, the ordered structure of the gels was kept. This was further verified from the AFM observation for the EDGA gel after mixing with AgNO3 for 4 h, as shown in Figure 1c. On the other hand, when the final DWSNTs or MWSNTs were formed, a similar FT-IR spectrum as that of the EDGA gel/AgNO3 was observed. In addition, a weak band appeared at 1729 cm-1, suggesting that some of the carboxylic acid formed a hydrogen bond again because of the reduction of the AgI to silver. However, most of the EDGA molecules still coordinated with the formed silver wall. This indicated that the organized structures are basically kept and EDGA was incorporated between (27) Nakamoto, K. Infrared and Raman Spectra of Inorganic and Coordination Compounds; Wiley: New York, 1986. (28) (a) Ohe, C.; Ando, H.; Sato, N.; Urai, Y.; Yamamoto, M.; Itoh, K. J. Phys. Chem. B 1999, 103, 435. (b) Hu¨hnerfuss, H.; Neumann, V.; Stine, K. J. Langmuir 1996, 12, 2561.

Synthesis of Double- and Multiwall SilVer Nanotubes

Langmuir, Vol. 22, No. 2, 2006 779

Figure 6. Possible mechanisms of the template synthesis of the DW and MWSNT under photoirradiation in a 1:1 mixture of ethanol/water. (a) Bolaamphiphilic nanotube. (b) Formation of a DWSNT. (c) Self-assembly of the bolaamphiphilic EDGA molecules and AgI on the surface of the formed DWSNT. (d) Formation of a three-layered SNT.

the nanotubes. Even after reduction, which was verified from the spectral change in the UV-vis spectra, the EDGA was still coordinated on the silver wall. This is similar to the self-assembly of carboxylic acid on the silver surface.29 We have tried to dissolve the EDGA molecules using organic solvent. However, it was found to be unsuccessful because of the coordination of the compounds with the silver wall. 4. Possible Mechanisms of Formation of the DWSNTs and MWSNTs. EDGA helical nanotubes were suggested to be rolled from a monolayer lipid membrane of EDGA with a thickness of ∼3 nm, in which both of the outer and inner surfaces of the EDGA nanotubes are hydrophilic.25 When the gel was mixed with AgNO3, the silver(I) cations can be coordinated with the carboxylic acid group, as shown in Figure 5b. When the carboxylic groups of the helical nanostructure were bonded with the silver(I) cation, the helical structure was destructed. There are two kinds of hydrogen bonds between the EDGA molecules. One is the hydrogen bond between carboxylic acid, the other is the hydrogen bond between the amide groups. Although the hydrogen bond between the carboxylic acid was destroyed after being mixed with AgNO3, the hydrogen bond between amide groups was still kept. Therefore, the ordered nanotube structure of the EDGA was kept even when AgI cations were reacted with the carboxylic acid groups (also verified from AFM in Figure 1c). Owing to the reducing ability of both light and ethanol, AgI could be reduced to a silver atom. The photoreduced silver atoms aggregate into single crystalline silver walls on the template. It should be noted that the slow reduction of the AgI cation to Ag is also important to grow the silver wall. When we used NaBH4 as a reductant, we only obtained a black deposit quickly in the test tube. When the gel/AgNO3 was irradiated longer, some of the EDGA molecules dissolved in the solution. Because of the affinity between silver and the electronegative carboxylic acid group, one end of the bolaamphiphilic EDGA could be coordinated on (29) (a) Tao, Y. T. J. Am. Chem. Soc. 1993, 115, 4350. (b) Tao, Y. T.; Huang, C. Y.; Chiou, D. R.; Chen, L. J. Langmuir 2002, 18, 8400.

the formed silver nanotube wall, as shown in Figure 6b. This process is very similar to the self-assembly of amphiphiles on silver surfaces. Because there is still another end of carboxylic acid, the excess AgI in the solution can be coordinated to the another end of carboxylic acid assembled on the silver surface. Subsequent reduction caused the formation of the third silver wall. In a similar manner, the fourth and fifth wall of silver could be formed. In forming the MWSNTs, because of the incomplete coordination of a third layer on the silver wall, there were some defects in the MWSNTs.

Conclusions DWSNTs and MWSNTs could be synthesized by using the uniform EDGA nanotubes as the templates. During the formation of the silver wall nanotubes, two steps are important. Silver cations were first aligned onto both the inner and outer surfaces of the helical EDGA nanotubes through the coordination during which the helical structure of the EDGA nanotubes was destroyed, while the ordered nanotube structures were kept. After reduction of the AgI cation, the DWSNTs were formed. The silver layer was separated by a distance of 2.7 nm, which is just the molecular length of EDGA. Both the dissolved EDGA molecules and AgI cations could be further self-assembled on the formed DWSNT surface. Subsequent reduction caused the formation of the third, fourth, and fifth wall of the silver nanotubes. The relative molar ratio of EDGA to the silver(I) cations was an important factor for the formation of well-defined DWSNTs and MWSNTs. Less AgI cations only provided the nanoparticles, while the excess of AgNO3 caused the aggregation of silver nanoparticles on the DWSNTs. The results opened a new way to synthesize the multiwall metal nanotubes in a controlled way. Acknowledgment. This work was supported by the NSFC (numbers 20273078, 90306002, and 20533050) and the fund from the Chinese Academy of Sciences. LA0517787