18    EXPERIMENTAL STUDY OF TOXIC METAL—
SORBENT REACTIONS IN A
BENCH SCALE COMBUSTOR
Timothy M. Owens, Graduate Assistant
Pratim Biswas, Professor
Aerosol and Air Quality Research Laboratory
Department of Civil and Environmental Engineering
University of Cincinnati
Cincinnati, Ohio 45221-0071
INTRODUCTION
Toxic metals may enter a combustion chamber in many physical or chemical forms, for example, as a consitituent of a hazardous or municipal solid waste to be incinerated or as a trace quantity in coal. Control of toxic metal emissions from combustors is currently being stipulated by
the U. S. EPA in the form of maximum achievable control technologies (MACT) for 11 metals
(As, Be, Cd, Co, Cr, Hg, Mn, Ni, Pb, Sb, and Se) and their compounds under Title III of the 1990
Clean Air Amendments. However, this is a problematic proposition for volatile metals such as
mercury, lead, and cadmium. Volatile metals introduced into a combustion device may enter the
gas phase at combustor temperatures, and then as the temperature decreases downstream of the
combustor, undergo nucleation to form a submicrometer sized aerosol that is not easily captured
by conventional air pollution control equipment.1"3 Several researchers have proposed using
bulk solid sorbents in several geometries (packed bed, fluidized bed, and dry sorbent injection)
and have demonstrated the potential to effectively remove metals from an air stream by means of
chemisorption.4-8 Chemisorption is preferred over physisorption because the metals may become
enviromentally immobile when chemisorbed by certain materials.7 Equilibrium calculations have
been performed for a number of toxic metals reacting with sorbents to establish the relative effectiveness of various sorbents.9
Sorbent injection into a combustion chamber can be effective for two reasons: (1) the metal
compound vapors can be scavenged by the sorbents in the combustion chamber thereby suppressing nucleation of the metal vapor compounds at the combustor exit and, (2) the metal compounds can be associated with a larger sized particulate mode which is captured more easily by
conventional particulate control devices. Although potentially effective, the use of bulk solid sorbents to chemisorb metal vapors is hindered by several physico-chemical considerations. As a
metals control method, injection of bulk solid sorbents into a combustor may encounter mass
transfer limitations due to outer surface reactions producing a metal-sorbent complex which
blocks the inner pore volume.6 Specifically, for lead capture by a kaolinite sorbent, the metal-
sorbent complex has a lower melting point than the sorbent resulting in the formation of a glassy
surface8 which inhibits further chemisorption. For the packed bed of sorbent, the bed must be
maintained at a relatively high temperature that is favorable for the metal-sorbent reaction. For
this reason, the packed bed strategy for metals removal by chemisorption would incur a significant additional energy cost. For the fluidized bed of sorbent following a combustor, sufficient
residence time must be allowed for adequate mass transfer and like the packed bed, additional
energy is required to maintain the bed temperature. The fluidized bed strategy has an additional
disadvantage in that metal compounds exiting a combustor and entering a fluidized bed system
may be in the form of a submicrometer sized aerosol (due to nucleation of the metal compound
vapor) and the fluidized bed system is not particularly effective in removing this fine particulate
mode. However, sorbent injection into a fluidized bed combustor may enhance metal compounds
capture downstream of the bed by providing additional surface area for condensation of the
50th Purdue Industrial Waste Conference Proceedings, 1995, Ann Arbor Press. Inc., Chelsea, Michigan 48118.
Printed in U.S.A.
173