Catalysis of Nanotube Growth

Carbon nanotubes (CNTs) represent a holy grail in the field of materials engineering. However, many characteristics that cannot currently be well manipulated need to be made controllable in order to improve the practicality of these materials. The properties of CNTs vary with diameter, helicity, and thickness (or walledness). Although in individual nanotubes these properties are easily isolable, the ability to create bulk quantities of highly controlled CNTs has not yet been developed.

The focus of this research is to gain a better understanding of the mechanisms of nucleation and growth, especially as concerns the yield of these syntheses. To this end, we employ computational methods to simulate what occurs in a physical synthesis of CNTs. In particular, we model interactions between metals and carbon chiefly using a reactive force field method, ReaxFF, wihch in turn is parametrized from DFT data.

Initially, we compare the interactions of Ni, Co, and Cu atoms with carbon. Ni and Co are known to be effective catalysts, whereas Cu is known to exhibit poor catalytic activity. Therefore, by discovering the differences in their reactivity, it should be possible to determine the most important factors present in a high-yielding CNT catalyst. Although these dynamics calculations do not yet demonstrate full scale nanotube synthesis, we do observe rapid five- and six-membered ring formation with Co and Ni, as well as ready dissociation from cage formation, allowing the catalysts to move on from the already formed regions of the CNT and further catalyze ring formation. Cu, on the other hand, does not exhibit any ring formation, most likely due to stronger metal-carbon interactions. (Kevin D. Nielson, Adri C. T. van Duin, Jonas Oxgaard, Wei-Qiao Deng, and William A. Goddard III J. Phys. Chem. A 2005, 109, 493-499)

Subsequently, we are looking at the role of Ni and Ru Catalyst particles. Periodic DFT calculations show that the formation of a C-C between two adsorbed C atoms is 8.9 kcal/mol downhill on Ni111 and 11.6 kcal/mol uphill on Ru111. This corresponds with, and explains why Ni is experimentally known to catalyze CNT growth, while Ru will not readily accomplish this. ReaxFF dynamics demonstrate the implications of this key energetic difference. Dynamics run on a Ni carbide particle initially configured to have no C-C bonds show the formation of C-C bonds and ring structures on the particle surface, which are the initial stages of CNT nucleation. In contrast, dynamics run on a Ru particle which initially has multi-C structures on the surface, shows Ru breaking up the C-C bonds to form a carbide phase.

Personnel: Sinchul Yeom, Jonathan E. Mueller
This project is directed by Dr. Adri van Duin.