Page 44 - Adsorbents fundamentals and applications

TEMPERATURE SWING ADSORPTION AND PRESSURE SWING ADSORPTION 29

Chapter 5, Yang, 1987). Mass and heat transfer are considered dispersive forces,
which have a dispersive or smearing effect on the concentration and temperature
fronts. The detailed bed profiles and breakthrough curves can be calculated only
by the numerical solution of the mass and heat balance equations (Yang, 1987),
coupled with the equations for equilibrium adsorption from mixture.
Experimental and theoretical results on the TSA process focused on the first
cycle of adsorption, followed by regeneration by using hot gas purge (Pan and
Basmadjian, 1970; Basmadjian et al., 1975a and 1975b; Carter, 1975). Because
the bed regenerated by hot purge is usually very clean, the cyclic steady state is
approached in the first cycle. This fact is implied but not clearly stated in several
early studies where the cyclic experiments were performed but only the first-cycle
results were reported (Carter, 1975; Chi and Cummings, 1978). This finding is
in contrast with the inert (cold) purge cycles, where several cycles are needed
to approach a cyclic steady state (Bunke and Gelbin, 1978). The difference is
due to the fact that cold purge is relatively ineffective and that the TSA cycle
requires the times for adsorption and regeneration to be equal in length. Using
hot-air purge for n-hexane-activated carbon, Davis and LeVan (1989) showed
that cyclic steady state was reached after 2 or 3 cycles.
Regeneration by steam is more practical, but more difficult to understand
(Scamehorn, 1979; Wankat and Partin, 1980; Schork and Fair, 1988; Schweiger
and LeVan, 1993). For steam regeneration, one needs to consider the condensing
flow as well as the difficult problem of adsorption equilibrium for hydrocar-
bon–water mixtures. Schork and Fair (1988) observed a high-temperature middle
zone in the bed due to the heat of adsorption of steam. The system of n-
hexane-carbon-steam has been studied in detail by LeVan and co-workers. Three
plateau zones were observed in the regenerating bed. The middle zone, at an
intermediate temperature between the feed steam and the front zone containing
the initial gas in the bed, is where the regeneration actions occur. Schweiger
and LeVan (1993) described such actions as “steam distillation” of the water-
immiscible hydrocarbon. In this zone, steam adsorbs and condenses, causing the
fluid velocity to reduce to a near stop beyond this zone. They have also given
a complete model that could fully describe their experimental observations in
steam regeneration.
The regeneration step in the TSA cycle requires time to heat, desorb, and cool
the bed. It is often the time-limiting step in the TSA cycle and is also the most
complex and least understood one. The following discussion on TSA focuses on
the regeneration step and presents some simple rules for design.

Minimum Purge Temperature. The minimum purge temperature was derived
based on the equilibrium theory by Basmadjian et al. (1975a and 1975b), which
has been discussed in detail elsewhere (Yang 1987; Basmadjian, 1997).
Efficient desorption is accomplished at temperatures above the “characteristic
temperature,” T 0 . The characteristic temperature is equal to the temperature at
which the slope of the adsorption isotherm at the origin is equal to C ps /C pb ,the
ratio of the heat capacities of the solid phase and the inert carrier gas. For a