208g Facilitating Density-Based Electronic-Type Separation of Carbon Nanotubes Via Chemical Reactions: A Modeling Study

Nitish Nair1, Woo-Jae Kim1, Monica Usrey2, and Michael S. Strano3. (1) Department of Chemical Engineering, Massachusetts Institute of Technology, Building 66-580, 25 Ames St, Cambridge, MA 02141, (2) University of Illinois at Urbana/Champaign, 104 RAL, 600 S. Mathews Ave., Urbana, IL 61802, (3) 66-566 Department of Chemical Engineering, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, MA 02139

A structure-reactivity relationship for electron transfer reactions of single walled carbon nanotubes (SWNT) has been derived and experimentally validated using 4-hydroxybenzene diazonium as a model electron acceptor. The formalism provides a mechanistic insight into electronically selective reactions. We have shown that covalently attached chemical groups can alter the densities of individual SWNT in a predictable and highly controllable manner, thus facilitating separation via ultracentrifugation. A hydrodynamic model is used to describe the motion of surfactant-suspended single walled carbon nanotubes in a density gradient, while being subjected to a centrifugal field. The number of surfactant molecules and functional groups adsorbed on each nanotube determine its effective density, and hence, its position in the gradient after centrifugation has been completed. Analysis of the spatial concentration distributions of unreacted CoMoCAT nanotubes suspended with 2 w/v% sodium cholate yielded 2.21, 2.36 and 2.74 surfactant molecules adsorbed per nanometer along the length of the (6,5), (7,5) and (8,7) nanotubes, respectively. A volume additivity model based is able to estimate the density difference between 4-hydroxy phenyl functionalized and non-functionalized HiPco SWNT as approximately 98.3 kg/m3, compared with 97.9 kg/m3 measured from density gradient centrifugation. We therefore conclude that chemical functionalization can provide an effective handle to separate out a particular SWNT from a typical diameter distribution.