Page 417 - Adsorption by Powders and Porous Solids

CHAPTER 12. PROPERTIES OF SOME NOVEL ADSORBENTS 407
clear, but it seems likely that the chromatographic results are in part due to a very
high degree of interconnectivity within the pore structure of AX21.
In this book we have generally adopted the convention of expressing the adsop
don as the amount adsorbed by unit muss of the outgassed adsorbent. From a techno-
logical standpoint, however, it is sometimes more important to consider the amount
adsorbed per unit volume of the adsorbent. It is evident that when this change is made
for a powder of low bulk density such as AX21 (bulk density = 0.3 g ~rn-~) the avail-
able adsorbing capacity is much less impressive (Sing, 1989). For this reason, for a
particular purpose it may be necessary to employ an adsorbent of moderate activity
rather than one of very high specific activity.

12.2.2. Activated carbon fibres and carbon cloth
The first high-strength carbon fibres were produced in the 1950s (see Donnet and
Bansal, 1984). The early carbonized products were rayon-based, but it was soon
found that the mechanical properties and the carbon yield could be improved by the
use of polyacrylonitrile (PAN) as the precursor. Also, less expensive fibres of
somewhat lower strength and modulus could be made from various other precur-
sors including petroleum pitch and lignin. However, cotton and other forms of
natural cellulose fibres possess discontinuous filaments and the resulting mechani-
cal properties were consequently found to be inferior to those of the rayon-based
fibres.
It was not long before the first activated carbon fibres (ACFs) were developed. In
the work of Economy and Lin (1971, 1976) highly porous carbon fibres were pre-
pared from .Kynol, a fibrous phenolic precursor. Carbonization was carried out in
nitrogen at 800°C and activation occurred in steam at 750-1000°C. The products
appeared to be predominantly microporous and were found to be effective for the
removal of low levels of certain pollutants (e.g. phenol and pesticides) from air or
aqueous solutions.
Another important development was the disclosure by Bailey and Maggs (1972;
Bailey et al., (1973) of a novel procedure for the manufacture of 'charcoal cloth'. The
continuous process developed in the laboratories of the Chemical Defence
Establishment, Porton Down, England, involved three main stages: (1) immersion of
the roll of viscose rayon cloth in an aqueous solution of inorganic chlorides (e.g.
ZnC1,. AlCl, and NH,Cl); (2) oven drying in nitrogen; (3) carbonization and activa-
tion in carbon dioxide.
The Type I character of the nitrogen and toluene isotherms displayed in Figure
12.1 indicates that a typical sample of activated charcoal cloth produced by the orig-
inal Porton process had novel adsorbent properties. The activated material was strong
and flexible and had a BET area of over 1200 m2 g-', a wide distribution of micro-
pores (no detectable mesoporosity) and a small external surface area (Atkinson et al.,
1982; Hall and Williams, 1986). In view of its early promise, it was logical to attempt
to control the pore structure of charcoal cloth. A systematic study of the development
of porosity was therefore undertaken by Atkinson et al. (1982, 1984) and Freeman
et al. (1987-1991).