The Coordination Chemistry of Carbon Nanotubes: a Density Functional Study through a Cluster Model Approach Francesca Nunzi, Francesco Mercuri, and Antonio Sgamellotti* Dipartimento di Chimica e Istituto di Scienze e Tecnologie Molecolari ISTM CNR, UniVersita ` di Perugia, 1-06123 Perugia, Italy Nazzareno Re Facolta ` di Farmacia, UniVersita ` G. D’Annunzio, 1-66100 Chieti, Italy ReceiVed: May 8, 2002; In Final Form: July 24, 2002 Polycyclic aromatic hydrocarbons (PAHs) have been employed as models to investigate the functionalization of SWNTs sidewalls with transition metal complexes at the NL-DFT level, pointing out the most favorable coordination sites and the electronic properties of the resulting system. Molecular fragments topologically resembling the honeycomb lattice have been chosen, i.e., the pyrene (C 16 H 10 ) and the coronene (C 24 H 12 ) to respectively investigate an η 2 interaction on the CsC bond and an η 6 interaction on the hexagonal ring with suitable transition metal complexes. To reproduce the curvature of both (n,0) and (n,n) SWNTs surface, constrained geometry optimizations have been performed on these systems. Both geometrical parameters and bonding energies for the (PH 3 ) 2 M(C 16 H 10 ) (M ) Ni, Pt) complexes suggest that η 2 bonding of metal fragments to nanotubes is much weaker than to fullerene, making questionable the stability of η 2 M(PH 3 ) 2 complexes of carbon nanotubes. The η 2 metal interaction with (9,0) and (5,5) bent-constrained pyrene models do not show any dependence on the (n,m) chiral vector. The analysis of the Cr(CO) 3 (C 24 H 12 ) complexes shows that the (9,0) and (5,5) bent-constrained coronenes, modeling carbon nanotubes, enforce the chromium atom to arrange an η 2 or η 4 coordination on the hexagonal ring, rather than an η 6 coordination, that is more favorable for the planar aromatic systems of benzene and planar coronene, modeling graphene sheets. 1. Introduction Single-wall carbon nanotubes (SWNTs) 1 have induced great research interest because of their unusual physical, chemical, and mechanical properties. 2-5 These unique characteristics make them ideal candidates as building blocks of molecular scale machines and nanoelectronic devices. Understanding the chem- istry of SWNTs is a viable route to developing controlled synthesis methods, enhance their solubility, and make them more amenable for the assembling of nanostructure precursors. The functionalization of the sidewalls with transition metal atoms, metal surfaces, or metal compounds provides new prospects in the manipulation of the electronic and magnetic properties of carbon nanotubes. The metal coordination represents an essential tool in organometallic crystal engineering, because it allows us to manipulate the functionalized systems and to modify the properties of the ligands by varying the nature of the metal and co-ligands, the coordination geometry, and the oxidation states. 6 The difficulties in obtaining pure SWNTs and in dissolving them in solvents make the chemical modification of SWNTs sidewalls a relatively unexplored area of research. SWNTs have been fluorinated, 7 and subsequently, fluorine atoms have been displaced with alkanes. 8 Chen et al. reported the derivatization of SWNTs dissolved in organic solutions with thionyl chloride and octadecylamine, as well as with chlorine through the use of dichlorocarbene. 9 Moreover, SWNTs have been ultrasonicated in a monochlorobenzene solution of poly(methyl methacry- late), 10 and they have been noncovalently functionalized using a bifunctional molecule to form amide bonds for protein immobilization. 11 However, not very much has been adressed in the area of the coordination of SWNTs with transition metals. Only recently, Banerjee and Wong accomplished the synthesis of functionalized carbon nanotubes by reaction with a transition metal complex, trans-IrCl(CO)(PPh 3 ) 2 . 12 Spectroscopical analy- sis suggests that the iridium atom is coordinated to carbon nanotubes in an η 2 fashion, analogously to electron-deficient alkenes, whereas oxidized nanotubes allow the metal coordina- tion through the oxygen atoms coating the sidewalls. Another interesting topic is the addition of Cr(CO) 3 metal complexes to the six-membered rings of SWNTs. 13 Among the inspected transition metal interactions with π systems, 14,15 (η 6 -arene)tri- carbonyl-chromium complexes have been extensively studied, 16-18 discovering a wide range of applications, such as chiral building blocks in total synthesis and their involvement in ligand design for asymmetric catalysis. 19-24 Indeed, it has been shown that the complexation with Cr(CO) 3 significantly modifies the structure and the reactivity of the complexed arenes. 25,26 Although Wilson 13 attempted the reaction of Cr(CO) 6 with SWNTs, pointing out the formation of an adduct “SWNT- Cr(CO) 3 ”, experimental difficulties in the manipulation of the nanotubes made ineffective the characterization of the reaction product. A detailed theoretical account for the interactions between transition metal complexes and carbon nanotubes is currently lacking, representing a challenging task for the theoretical * To whom correspondence should be addressed. E-mail: sgam@ thch.unipg.it. Phone: +39 075 5855516. Fax: +39 075 5855606. 10622 J. Phys. Chem. B 2002, 106, 10622-10633 10.1021/jp026088f CCC: $22.00 © 2002 American Chemical Society Published on Web 09/19/2002