No abstract

A procedure, combining molecular simulation, Raman spectroscopy, and standard nitrogen adsorption, is developed for structural characterization of single-walled carbon nanotube (SWNT) samples. Grand canonical Monte Carlo simulations of nitrogen adsorption are performed on the external and internal adsorption sites of homogeneous arrays of SWNTs of diameters previously determined by Raman spectroscopy of the sample. The results show the importance of the peripheral grooves of a nanotube bundle at low relative pressure and the insensitivity of nanotube diameter toward adsorption on the external surface of the bundle at higher pressures. Simulations also reveal that samples containing thin nanotubes have less internal adsorption capacity that saturates at lower pressure than those comprising large diameter nanotubes. The fraction of open-ended nanotubes in a sample can be estimated by scaling the simulated internal adsorption inside nanotubes to obtain a near perfect fit between simulated and experimental isotherms. This procedure allows extrapolation of adsorption properties to conditions in which all nanotubes in the sample are open-ended.

No abstract

No abstract

Hexane adsorption on single-walled carbon nanotube (SWNT) bundles is studied by both simulation and experimentally using a previously developed computer-aided methodology, which employed a smaller physisorbed probe molecule, nitrogen, to explore the porosity of nanotube samples. Configurational-bias grand canonical Monte Carlo simulation of hexane adsorption on localized sites of the bundles is carried out to predict adsorption on their external surface and in their internal sites. These localized isotherms are then combined into a global isotherm for a given sample by using knowledge of its tube-diameter distribution and structural parameters, such as the fraction of open-ended nanotubes and the external surface area of bundles in samples, which have been independently determined from the standard nitrogen adsorption isotherm. The near-perfect replication of experimental isotherms demonstrates the validity of our method for structural characterization of SWNT samples. The effect of temperature on adsorption is also studied and the simulation results are extrapolated to predict the limiting hexane adsorption capacity of the samples. The similarity between the hexane adsorption isotherms and those of other organic molecules demonstrates that the adsorption mechanisms explored here are not specific to hexane, and that the proposed methodology can be potentially applicable to other sorbates with equal success.

A practical approach for adsorption modeling of heterogeneity of single-walled carbon nanotube (SWNT) bundles has been developed. The method integrates experimental analysis with grand canonical Monte Carlo (GCMC) simulation of a small probe molecule, such as nitrogen at 77 K. Using this method, it is possible for one to separately estimate adsorption inside the nanotubes, adsorption on the external surface of the bundles, and adsorptive contributions from the impurities present in samples. By introducing a scaling parameter for adsorption in the internal porous volume of the bundles, the predicted adsorption isotherm results in a near replication of the experimental N 2 adsorption isotherm. We refer to this parameter as the volume fraction of open-ended nanotubes. Our GCMC-assisted experimental characterization method has been applied successfully to several commercial samples obtained from different suppliers, such as MER Corp., Carbon Nanotechnologies Inc., Carbon Solutions Inc., Carbolex Inc., and BuckyUSA. It was found that the volume fraction of openended SWNTs in these samples ranged between 0 and 55%. The majority of the samples were subjected to some purification treatment by the manufacturer and exhibited an already high BET surface area of hundreds of square meters per gram. The near-perfect reproduction of the experimental N 2 (77 K) adsorption isotherm for each of the tested samples shows that our characterization method is not specific to a particular sample and can be extended to most SWNTs successfully. The fraction of open-ended SWNTs cannot otherwise be estimated by visual characterization of the samples because of the large aspect ratio of nanotubes and the spaghetti-like arrangement of the bundles. Our method has the potential to become a standard technique to quantify this structural property of SWNT samples.

We are reporting gravimetric measurements of adsorption isotherms and kinetics of water vapor in singlewalled carbon nanotubes (SWNTs). Adsorption was facilitated in an open configuration. Adsorption capacities of SWNTs were found to be approximately one-half of those of activated carbon and activated carbon fiber. The water adsorption isotherms of SWNTs followed type V characteristics, which are typical for a surface chemistry mediated adsorption of water. The isotherm models, the Dubinin-Serpinski equation and the Do equation, were fitted to the SWNT isotherms. The information about the SWNT surface chemistry could be extracted from data fitting to these equations and was found to be consistent with the same observed from Raman scattering of the SWNTs. The adsorption and desorption rate constants were calculated for all SWNT samples as a function of relative pressure using the linear-driving force model commonly used for activated carbons. The rate constants for both carbon types were in the general range of 1 × 10 -3 to 3 × 10 -3 s -1 with values decreasing with increasing vapor pressure; however, the distinction between adsorption on the primary sites versus capillary condensation was not apparent for SWNTs. This work can provide descriptive experimental data (detection limit ) 0.1 µg) to aid molecular simulation studies of water adsorption in microporous and nanoporous carbon adsorbents.

No abstract