Page 264 - Adsorbents fundamentals and applications

CARBON NANOTUBES 249

52 meV for Ne. They actually calculated the isosteric heats of adsorption (q st )
from the temperature dependence first (at low loadings), and obtained the binding
energies (ε)by q st =−ε + 2kT,where k is the Boltzmann constant. From the
ε values above, the heats of adsorption were approximately 23 kJ/mol for CH 4 ,
29 kJ/mol for Xe, and 7.4 kJ/mol for Ne. The SWNT bundles were reported as
having closed ends. The results were interpreted as adsorption occurring on the
outer surfaces and “ridges” of the bundles, not in the spaces between the tubes
within the bundles. Simulation of Xe on the outer surfaces of SWNTs has been
studied by Stan and Cole (1998), who concluded that the binding of Xe on the
outer surface was weaker than that on planar graphite by about 20%. However,
their estimated binding energy of 22.6 kJ/mol was close to the experimental
data of Talapatra et al. (2000). Kuznetsova et al. (2000) measured the adsorption
of Xe on both closed and open SWNTs at 95 K, using TPD. The desorption
activation energy of Xe from saturated phase inside the tubes was 26.8 kJ/mol.
This value was also in reasonable agreement with that obtained by Talapatra
et al. and Stan and Cole. In a study of Muris et al., (2000), adsorption isotherms
of CH 4 and Kr at 77–110 K were measured on closed bundles of SWNTs, with
a tube diameter of 13.7 ˚ A and intertubular distance of 17 ˚ A. The isotherm of
CH 4 on SWNTs had two steps, with isosteric heats of adsorption of 18.3 and
11.1 kJ/mol for the first and second steps, respectively. The heat of adsorption
of the first step of adsorption for CH 4 on planar graphite was 14.9 kJ/mol. Their
values seemed to be substantially lowered than the others. An earlier measurement
of adsorption of CH 4 on MWNTs was performed on large tubes, ranging from
10 nm to 2 µm diameter (Mackie et al., 1997). In this study, typical wetting and
capillary condensation behavior was seen.
An interesting result from the study of Kuznetsova et al. (2000) involved
the opening of the tubes by heat-treatment under high vacuum. Evolution of
CH 4 ,CO, H 2 ,and CO 2 gases began when heating the SWNTs to temperatures
◦
◦
above 300 C. Their final temperature was 800 C. After such heat-treatment, the
saturated amount of Xe increased by a factor of 23, indicating opening of the
tubes. They postulated that the surface functionalities such as −COOH blocked
the entry ports for adsorption at the nanotube ends and at defect sites on the
walls. The thermal destruction of these functionalities would lead to opening of
the pores. A similar postulation was actually made previously by Rodriguez and
Baker (1997) from their studies of hydrogen storage on GNFs. This important
point will be further discussed in detail in Chapter 10.

Isotope Separation. Separation of isotope mixtures (e.g., separations of tri-
tium and deuterium from hydrogen) is a difficult task that requires energy-
intensive processes such as diffusion, chemical exchange and laser isotope separa-
tion. Adsorption processes (mainly by chromatography) have attracted continuing
interests; however, good sorbents have yet to be found.
Commercial sorbents have been considered for isotope separations. These
include: D 2 -H 2 on alumina at 77.4 K (Katorski and White, 1964; King and Ben-
son, 1966), D 2 -DT-T 2 on 13X zeolite at various low temperatures (Maienschein