Magnetic permeability profiles in adsorption beds



Journal Title

Journal ISSN

Volume Title



The rate of adsorption of an adsorbate in a fluid flowing through a bed of adsorbent depends on kinetic and mass transfer effects. These phenomena cause a dispersion of the adsorbate species, both in the fluid and on the solid. The proper design of a system to remove efficiently and effectively these species requires a knowledge of the mechanism of the adsorption. With such information, dispersion may be minimized, yielding maximum utilization of the adsorbent. Numerous mathmatical models have been proposed for the movement and dispersion of the adsorption wave. These models are either theoretical with adjustable parameters or empirical for a particular system. The parameters are usually obtained from effluent breakthrough curves. Multiple analyses with variable length beds are necessary to determine the dispersion with length. In 1946 Thiele commented that "theory is ahead of experimental work in this field". The theory has been developed considerably since then, but experimental techniques are essentially unchanged. Recently, Richardson proposed a method for determining adsorption profiles for adsorbents that change magnetically upon adsorption. The voltage induced into a coil of wire in a changing magnetic field is proportional to the permeability and the amount of material inside the coil. By moving a bed of magnetic material into the coil, the change in the voltage induced into the coil with the position of the bed relative to the coil should be directly proportional to the change in the concentration of the material entering the coil. However, a real bed induces a voltage into the coil in a non-linear manner while approaching the coil and before actual entrance, causing "end effects". This thesis describes a technique for determining the "end effects" and accounting for them mathmatically to render adsorption data for realistic adsorbate profiles. Although considerably more complex than Richardson's "ideal coil" method, the technique developed will describe a two parameter equation relating dispersion of the concentration wave to its position in the bed. From magnetic measurements, the validity of models can be evaluated as the adsorbate wave moves down the bed, thus reducing the number of beds required by previous methods. An improvement of this technique is suggested for obtaining a third parameter which would provide sufficient data to reject a proposed model with measurements for a single profile.