The injection of calcium-based sorbents into coal-fired boilers for reaction with, and reduction in the levels of, sulfur dioxide (SO,) in the flue gas has undergone considerable research and development. Significant effort has also been made in developing models for the overall reaction CaO + SO, + 1/2 O2 -CaS0, in order to better predict the effects of system and sorbent variables upon performance. Development of internal surface area (Borgwardt and Bruce, 1986) and pore structure (Hartman and Coughlin, 1974) is necessary for measurable reaction to occur. Further, since the calcium sulfate (CaSO,) product has a larger molar volume than the reactant, calcium oxide (CaO), the extent of pore volume development may control the levels of reaction. The molar volume of CaSO, is related to its density. One of the earlier references (Hartman and Coughlin, 1974) to the CaS04 product density uses the molar volume value of 52.2 cm3/mol or 2.6 g/cm3. This leads to a product/reactant expansion ratio, z = vCaSo,/VCaO 7 of 52.2/16.9, or 3.09. Use of this particular handbook value (Weast, 1968) of density (or its associated Z value) has continued in subsequent papers (Hartman and Coughlin, 1976;Ramachandran and Smith, 1977;Hartman et al., 1978;Bhatia and Perlmutter, 1981; Sotirchos and Yu, 1985;Reyes and Jensen, 1987;Borgwardt et al., 1987;Kocaefe et al., 1987;Simons and Garman, 1986;Yu and Sotirchos, 1987). One instance (Hartman and Trnka, 1980) is noted where the value of 2.97 g/cm3 (45.8 cm3/mol) has been used, cited from an unidentified *Correspondence concerning this paper should be addressed to B. K. Gullett handbook. This last case results in a Z value of 2.7 1. Apparently the earliest documented reference questioning the commonlyaccepted molar volume value is by Dam-Johansen (1987). This work determined a molar volume of 48.06 cm3/mol from a model comparison with experimental porosity/conversion data. Deviation from the inferred 46.0 cm3/mol (2.96 g/cm3) value was attributed to experimental errors and an insufficient model.The value of 2 is significant particularly for modeling the sulfation reaction. Since the larger volume product fills or blocks sorbent pore volume, slowing or stopping the reaction, the ability of models to appropriately simulate laboratory reaction results is contingent upon proper determination of the physical parameters involved. To this end, analyses to identify the form of the CaSO, product and determine its density (molar volume) were performed.
Experimental MethodSamples of calcium hydroxide [Ca(OH),] and calcium carbonate (CaC03) were reacted at 80OOC for 5 min supported on quartz wool in a fixed-bed reactor. Preheated reactor gas [0.3% SO,, 5% oxygen (02), balance nitrogen (N2)] at 23 Ljmin and standard temperature and pressure (STP) passed through the sample bed, converting the sample to the reactive CaO form and then partially reacting it to form CaS04. (Further details of similar experimental procedures are available in Borgwardt and Bruce, 1986.) The samples were analyzed by x-ray diffra...