Ar, CCl 4 and C 6 H 6 adsorption outside and inside of the bundles of multi-walled carbon nanotubes—simulation studyw Sylwester Furmaniak, a Artur P. Terzyk,* a Piotr A. Gauden, a Rados$aw P. Weso$owski a and Piotr Kowalczyk b Received 2nd December 2008, Accepted 4th March 2009 First published as an Advance Article on the web 6th April 2009 DOI: 10.1039/b821633a This is the first paper reporting the results of systematic study of the adsorption of Ar, C 6 H 6 and CCl 4 on the bundles of closed and opened multi-walled carbon nanotubes. Using grand canonical Monte Carlo (GCMC) and molecular dynamics (MD) simulations, we also study the effect of the introducing defects in the external and internal walls of osculating and separated nanotubes on Ar diffusion and on adsorption of all three adsorbates. The Ar diffusion coefficients obtained are very sensitive to the presence of defects. Simulated isotherms are discussed to show the relation between the shapes of the high resolution a s -plots and the mechanisms of adsorption. From obtained data, as well as from geometric considerations, from the VEGA ZZ package, and from simulations (ASA), the values of surface areas of all nanotubes are calculated and compared with those obtained using the most popular adsorption methods (BET, a s and the A,B,C-points). We show that the adsorption value for the C-point of the isotherm should be taken for the calculation of the specific surface area of carbon nanotubes to obtain a value which approaches the absolute geometric surface area. A fully packed monolayer is not created at the A-, B- or C-points of the isotherm; however, the number of molecules adsorbed at the latter point is closest to the number of molecules in the monolayer as calculated via the ASA method, the VEGA ZZ package or from geometric considerations. 1. Introduction The presence of various defects in single- and multi-walled carbon nanotubes (SWCNs and MWCNs) is well documented by STM, 1 ESR, 2 Raman 3 and TG 4 measurements and different defects are shown (by theoretical calculations) to be stable. 3,5 Recent HRTEM studies have shown direct and clear evidence for the presence of point defects like adatoms, monovacancies, interstitial-vacancy defects and pentagon–heptagon pairs in carbon nanostructures, and their evolution over time. 6 Those defects can be created at the preparation stage of carbon materials; however, to study their properties, they are usually introduced using the method of irradiation via an electron beam 6,7 or via protons or atomic ions. 8 Probably the simplest method of creating defects is nanotube purification, using, for example, chlorine oxidation. As it was shown by Yuan et al, 9 TEM images of MWCNs after reaction with chlorine water showed the presence of lattice defects on the external walls of nanotubes. Similar results are also observed during purification/ oxidation of nanotubes with nitric acid after desorption of surface oxygen functionalities. Adsorption on defects created during the preparation of nanotubes seems to be the most important effect dominating the electrical response of nanotube sensors (the defect sites form low-energy adsorption centres that also serve as nucleation sites at high vapour concentrations). The chemical sensitivity of sensors can be increased significantly by controllably introducing a low density of defects along the nanotube sidewall. 10 Carbon nanotubes with defects are also applied to forming all-carbon based microelectronic devices. 11 Creation of defects in SWNTs by ultrasonication with organic solvent can also be applied to the synthesis of fullerenes. 12 From this short introduction it can be seen that investigation of defects in carbon nanotubes is an important subject in current carbon nanotechnology. On the other hand, it is surprising that the studies of adsorption properties of carbon nanotubes with defects are rarely met in the literature. Based on a simulation experiment, we have recently shown that the assumption of the presence of defects in internal walls of carbon nanotubes can lead to an explanation of the ‘‘wavy’’ shapes of experimental high-resolution a s -plots. 13 Thus, in the present paper, we study how the appearance of defects on external and internal walls of MWCNs changes adsorption properties towards Ar, C 6 H 6 and CCl 4 . Those adsorbates were chosen because they are the most popular molecules for characterisation of the structural properties of carbon materials. By the grand canonical Monte Carlo (GCMC) simulation method we study the mechanism of adsorption and the influence of defects on this mechanism. The influence of a N. Copernicus University, Department of Chemistry, Physicochemistry of Carbon Materials Research Group, Gagarin St. 7, 87-100, Torun ´, Poland. E-mail: aterzyk@chem.uni.torun.pl; Web: http://www.chem.uni.torun.pl/Baterzyk/ Fax: (+48) (056) 654-24-77; Tel: (+48) (056) 611-43-71 b Applied Physics, RMIT University, GPO Box 2476V, 3001, Victoria, Australia w Electronic supplementary information (ESI) available: Simulated isotherms and enthalpy plots, simulation parameters, snapshots of simulations and Movies 1–8. See DOI: 10.1039/b821633a 4982 | Phys. Chem. Chem. Phys., 2009, 11, 4982–4995 This journal is  c the Owner Societies 2009 PAPER www.rsc.org/pccp | Physical Chemistry Chemical Physics