Page 335 - Adsorbents fundamentals and applications

320 SORBENTS FOR APPLICATIONS

catalyst, 17.4 wt % Ni and 1.9 wt % Mg. The BET surface area of this sample
2
was 184 m /g. The pure catalyst after the pretreatment had a hydrogen capacity,
due to MgO and its reduced forms. However, the hydrogen capacity of sample (A)
was far greater than that due to the residual catalyst. The catalyst-free MWNTs
(sample C) showed no hydrogen capacity. The hydrogen capacity of 0.65 wt %
for sample A was at 1 atm, hence was significant. This result is a clear indica-
tion that a dynamic process was at work that involved hydrogen dissociation and
spillover to the nanotubes. TPD runs at different heating rates were performed
on sample A. From the hydrogen desorption peak temperatures, the activation
energy for desorption was determined to be 23.3 kcal/mol. This value was close
to the bond energy of 27 kcal/mol for chemisorption of H atoms on the basal
plane of graphite, predicted from ab initio calculations (Yang and Yang, 2002).
To further elucidate the hydrogen uptake mechanism, TPD behaviors of the
sorbed hydrogen from two samples were compared: the Ni 0.4 Mg 0.6 O catalyst and
the MWNT with residual catalyst (Lueking and Yang, 2003). The results showed
◦
that the MWNT had an enhanced desorption peak at 140 C when compared
with the initial Ni 0.4 Mg 0.6 O catalyst, which had a dominant desorption peak at
◦
250 C. The dominant adsorption temperatures for these two samples were the
same. These results suggest that the adsorption process for MWNT is the same
as for the Ni 0.4 Mg 0.6 O catalyst, whereas desorption from the nanotubes occurs
directly from lower-energy carbon sites. Lueking and Yang (2003) also measured
hydrogen uptake of the MWNTs with residual catalysts at high pressures by
using the volumetric technique. The samples were pretreated in 1 atm H 2 at
◦
various temperatures (500–800 C), transferred to the high-pressure apparatus,
◦
followed by heating at 500 C in vacuo. The hydrogen uptake increased with the
◦
pretreatment temperature, being the highest at 800 C. Considerable amounts of
weight loss accompanied the pretreatments. Adsorption conditions were either at
a constant temperature at 69 atm, or under a temperature-pressure (T-P) cycle.
The T-P cycle consisted of a series of step changes where the temperature was
◦
◦
lowered as the pressure was increased: 122 C and 100 psia; 50 C and 500 psia;
◦
25 C and 1000 psia. After adsorption, desorption measurements were also made.
Higher hydrogen uptakes were obtained with the T-P cycle. The resulting uptake
amounts are summarized in Figure 10.30.
As discussed by Lueking and Yang (2003), the pretreatment not only removed
inactive materials but also activated the sample for hydrogen uptake. Other studies
have explained high temperature activation to be due to removal of chemisorbed
species (Park et al., 1999), destruction of surface functionalities that may block
pores (Kuznetsova et al., 2000), or graphitization of the nanotubes (Li et al.,
2001). However, these studies did not consider residual metal content, which has
been shown to affect the reactivity and gasification of nanotubes (Chiang et al.,
2001). The absolute hydrogen storage shown in Figure 10.30 is comparable with
results reported by others on MWNTs. At ambient temperatures, the reported
values range from 1.97% at 40 atm (Lee et al., 2002) to 4% at 100 atm (Li
et al., 2001) to 6.3% at 148 atm (Hou et al., 2002). Although comparison of the
results by Lueking and Yang (2003) to other hydrogen storage reports is not