(528f) Investigation of the Formation Mechanism of Single Walled Aluminogermanate Nanotubes

Nanotubular materials are important building blocks of a future nanotechnology based on synthesis of functional nanoparticles and their assembly into nanoscale devices with novel applications in areas such as electronics, biotechnology, sensing, separations, energy storage/management and catalysis. The discovery of carbon nanotubes[1] has generated a great deal of interest, and extensive research is presently being done on the synthesis, properties and applications of carbon nanotubes[2, 3]. However, the fundamental issues regarding how the nanotubes form are not well understood[4]. One of the primary reasons for the lack of information is due to the time scales at which carbon nanotubes are formed (around milliseconds), which makes it difficult to study their formation with any known characterization techniques[5]. An understanding of nanotube formation mechanisms would be invaluable in synthesizing new functional nanotube materials.

A suitable candidate for mechanistic study is the synthetic version of the nanotube mineral imogolite[5, 6]. Imogolite is a single-walled nanotube whose wall structure is identical to a layer of aluminum (III) hydroxide (gibbsite); with isolated silicate groups bound on the inner wall. The nanotubes are synthesized in water at mild conditions (pH 4, ~ 100°C) with a formation timescale of hours. An aluminogermanate analog (which is the focus of the present study) has also been successfully prepared by substitution of silicon with germanium in the synthesis solution[6]. The formation of the nanotubes could be either kinetically or thermodynamically controlled. Formation of nucleation sites in the early stage of the reaction and subsequent addition of oligomers and monomers as growth units to these sites indicates kinetic control. This mechanism would lead to a substantial increase in length of the nanotubes with synthesis time, given enough supply of reactants. On the other hand, thermodynamic control involves self-assembly of the monomers and oligomers into a short nanotube. In a recent study[7], we reported a systematic phenomenological study of growth of the aluminosilicate and aluminogermanate nanotubes over the period of 120 hrs. We found that the nanotubes do not grow in length significantly, but their concentration increases as a function of synthesis time.

Here we present a mechanistic study of the formation of the short (~ 15 nm), highly monodisperse aluminogermanate nanotubes. We divide the nanotube synthesis into five steps: (1) Hydrolysis ? The precursors are hydrolyzed to form a clear solution. (2) Basification ? Dilute base is added to adjust the pH of the solution to 5. (3) Acidification ? The solution pH is readjusted to 4.5 by adding dilute acid. (4) Equilibration ? The solution is allowed to equilibrate for approximately 3 hrs. (5) Reaction ? The solution is heated to 95°C for variable times (upto 400 hours in our study) under reflux conditions. We used dynamic Light Scattering (DLS) to track nanoparticle formation in solution during the various steps of the reaction. It is found that nanoparticle formation occurs only at the end of Step 2 (Basification) at solution pH of 5, but the nanoparticles formed are redissolved in Step 3 (as soon as the solution is reacidified) and do not reappear during equilibration. However, nanoparticle formation is observed again immediately after the onset of Step 5. TEM imaging of the products at each step of the reaction however, reveal no nanotube formation in Steps 1 to 4. These results indicate that the immediate precursors of the nanotube are small (sub-nanometer) particles. During Step 5, we observe a slow increase in the apparent nanotube length (measured by DLS) over a period of 400 hours, at various reaction temperatures in the region 25-95°C. We also perform two other types of experiments: (1) a semibatch experiment with inout of reactants throughout the synthesis, and (2) a seeded growth experiment wherein nanotubes are included in the reaction mixture at the start of Step 5. Combining these results, we analyze the relative roles of slow aggregation of nanotube materials and of nanotube growth by addition of precursors to the ends. We then use Diffusion Ordered Nuclear Magnetic Resonance Spectroscopy (DOSY NMR) to obtain information on both size and chemical environment of nanoparticles which are too small to analyze with DLS. Along with solid-state characterization (TEM, electron diffraction, and XRD) of the materials formed at various stages of reaction, the experiments provide a detailed picture of the synthesis of the aluminogermanate nanotubes that can allow the derivation of a quantitative mechanistic pathway of the nanotube formation. We discuss the implications of this study on the engineering of nanotube materials of controlled dimensions.

References: [1]Iijima, S., Nature, 1991. 354(6348): p. 56-58. [2]Snow, E.S., P.M. Campbell, and J.P. Novak, Applied Physics Letters, 2002. 80(11): p. 2002-2004. [3]Baughman, R.H., A.A. Zakhidov, and W.A. de Heer, Science, 2002. 297(5582): p. 787-792. [4]Sano, N., et al., Chemical Physics Letters, 2003. 378(1-2): p. 29-34. [5]Little, R.B., Journal of Cluster Science, 2003. 14(2): p. 135-185. [6]Wada, S. and K. Wada, Clays and Clay Minerals, 1982. 30(2): p. 123-128. [7]Mukherjee, S., V.A. Bartlow, and S. Nair, Chemistry of Materials, 2005. 17(20): p. 4900-4909.