Theoretical models have proposed that the nucleation and growth mechanism of carbon nanotubes (CNTs) has been affected by the catalytic activity of transition metals. The catalyst behavior during growth has been mainly associated as the responsible mechanism for the termination of CNT growth. Although several hypotheses have been developed to explain this mechanism, is still today an unresolved phenomenon. It was recently shown that the Ostwald ripening of iron (Fe) nanoparticles played a dominant role in the termination of CNT growth. The Ostwald ripening mechanism was further investigated as a function of thermal annealing in Hydrogen (H2) for iron (Fe) catalyst nanoparticles on various surfaces of aluminum oxide (i.e. sputtered alumina (a-Al2O3), and C- face sapphire (α-Al2O3)).
Experimental results showed that the growth kinetics of Fe nanoparticles on a-Al2O3 and α-Al2O3 obeyed the Ostwald ripening mechanism. The proportionality constant (K) for the kinetic equation of the Lyfshitz-Slyozov-Wagner (LSW) theory was calculated for a-Al2O3 and α-Al2O3, Ka-Al2O3 = 1.423 x 10-29 m3 s-1, and Kα-Al2O3 = 7.145 x 10-29 m3 s-1. The log-normal particle size distribution for the Fe nanoparticles was studied as a function of annealing time in H2 for 5, 10, and 15 minutes. It was found that the ripening of the particles is dependent on the surface, where a-Al2O3 < α-Al2O3. Then, after the growth of SWNTs, the particle size distribution of Fe was also investigated on C- and A- face sapphire surfaces. Experimental results showed that the Ostwald ripening rate of the catalyst nanoparticles is higher for the A- face than the C- face sapphire. The synthesis of SWNTs resulted in randomly oriented tubes on the C- face sapphire; as opposed to aligned tubes on the A- face sapphire surfaces. The SWNTs aligned along the specific crystalline directions corresponding to the anisotropic pseudo-1D array of Al atoms on the sapphire surface. The Ostwald ripening effect and the synthesis of SWNTs, was characterized by atomic force microscopy (AFM), scanning electron microscopy (SEM), and Raman Spectroscopy.
Due to experimental constraints, a theoretical prediction to modify the LSW theory was not developed at this instance. It is predicted that the particle size distribution will not agree with the LSW theory or with modified LSW equations, probably due to other conditions not included in the Ostwald ripening theory. This disagreement may be due to the existence of multiple and complex variables linked to the growth of CNTs, which are not taken into account in modern coarsening theories. Therefore, it is essential to further quantify all the associated variables that are responsible for the termination growth mechanism of CNTs. The development of a robust kinetic equation that can be coupled to experimentations would be necessary for the achievement of a more realistic Ostwald ripening model.